diff --git a/.coverage b/.coverage
new file mode 100644
index 0000000000000000000000000000000000000000..7bd62455cae2ec79af9cbc60f552afe43fca17d4
Binary files /dev/null and b/.coverage differ
diff --git a/.coverage.DESKTOP-ATMEKSV.24388.XANLHOVx b/.coverage.DESKTOP-ATMEKSV.24388.XANLHOVx
new file mode 100644
index 0000000000000000000000000000000000000000..e1f7e6f75f5aa9a7a33884193b67b7cc22082200
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.24388.XANLHOVx differ
diff --git a/.coverage.DESKTOP-ATMEKSV.29708.XilfwCcx b/.coverage.DESKTOP-ATMEKSV.29708.XilfwCcx
new file mode 100644
index 0000000000000000000000000000000000000000..c54e53aa1aa202b1377d33fea299950cda87d2bb
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.29708.XilfwCcx differ
diff --git a/.coverage.DESKTOP-ATMEKSV.33796.XdyCYWmx b/.coverage.DESKTOP-ATMEKSV.33796.XdyCYWmx
new file mode 100644
index 0000000000000000000000000000000000000000..81e53f831236a6e47a886ccb6403982ccf33b461
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.33796.XdyCYWmx differ
diff --git a/.coverage.DESKTOP-ATMEKSV.43284.XuuJaTEx b/.coverage.DESKTOP-ATMEKSV.43284.XuuJaTEx
new file mode 100644
index 0000000000000000000000000000000000000000..f723535d998644947972f91fa6e1bdbfa96b41ff
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.43284.XuuJaTEx differ
diff --git a/.coverage.DESKTOP-ATMEKSV.49024.XoHLhHmx b/.coverage.DESKTOP-ATMEKSV.49024.XoHLhHmx
new file mode 100644
index 0000000000000000000000000000000000000000..8c3bd15ceb242ea9c40c1b4be42b077d3d2db13b
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.49024.XoHLhHmx differ
diff --git a/.coverage.DESKTOP-ATMEKSV.50700.XPPOhAcx b/.coverage.DESKTOP-ATMEKSV.50700.XPPOhAcx
new file mode 100644
index 0000000000000000000000000000000000000000..fad3080e3e429bf14717ae3d448a946735a7c872
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.50700.XPPOhAcx differ
diff --git a/.coverage.DESKTOP-ATMEKSV.51388.XJOycQex b/.coverage.DESKTOP-ATMEKSV.51388.XJOycQex
new file mode 100644
index 0000000000000000000000000000000000000000..e1f7e6f75f5aa9a7a33884193b67b7cc22082200
Binary files /dev/null and b/.coverage.DESKTOP-ATMEKSV.51388.XJOycQex differ
diff --git a/.idea/.gitignore b/.idea/.gitignore
new file mode 100644
index 0000000000000000000000000000000000000000..13566b81b018ad684f3a35fee301741b2734c8f4
--- /dev/null
+++ b/.idea/.gitignore
@@ -0,0 +1,8 @@
+# Default ignored files
+/shelf/
+/workspace.xml
+# Editor-based HTTP Client requests
+/httpRequests/
+# Datasource local storage ignored files
+/dataSources/
+/dataSources.local.xml
diff --git a/.idea/bayesvalidrox.iml b/.idea/bayesvalidrox.iml
new file mode 100644
index 0000000000000000000000000000000000000000..fab03b6eca7d238814b681f1f06637c845bcc3f4
--- /dev/null
+++ b/.idea/bayesvalidrox.iml
@@ -0,0 +1,18 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<module type="PYTHON_MODULE" version="4">
+ <component name="NewModuleRootManager">
+ <content url="file://$MODULE_DIR$">
+ <sourceFolder url="file://$MODULE_DIR$/src" isTestSource="false" />
+ <sourceFolder url="file://$MODULE_DIR$/tests" isTestSource="true" />
+ </content>
+ <orderEntry type="inheritedJdk" />
+ <orderEntry type="sourceFolder" forTests="false" />
+ </component>
+ <component name="PyDocumentationSettings">
+ <option name="format" value="NUMPY" />
+ <option name="myDocStringFormat" value="NumPy" />
+ </component>
+ <component name="TestRunnerService">
+ <option name="PROJECT_TEST_RUNNER" value="py.test" />
+ </component>
+</module>
\ No newline at end of file
diff --git a/.idea/inspectionProfiles/profiles_settings.xml b/.idea/inspectionProfiles/profiles_settings.xml
new file mode 100644
index 0000000000000000000000000000000000000000..105ce2da2d6447d11dfe32bfb846c3d5b199fc99
--- /dev/null
+++ b/.idea/inspectionProfiles/profiles_settings.xml
@@ -0,0 +1,6 @@
+<component name="InspectionProjectProfileManager">
+ <settings>
+ <option name="USE_PROJECT_PROFILE" value="false" />
+ <version value="1.0" />
+ </settings>
+</component>
\ No newline at end of file
diff --git a/.idea/misc.xml b/.idea/misc.xml
new file mode 100644
index 0000000000000000000000000000000000000000..a6218fed0aeb0cbb03b46a9064efeeda11861bf6
--- /dev/null
+++ b/.idea/misc.xml
@@ -0,0 +1,7 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+ <component name="Black">
+ <option name="sdkName" value="Python 3.11" />
+ </component>
+ <component name="ProjectRootManager" version="2" project-jdk-name="Python 3.11" project-jdk-type="Python SDK" />
+</project>
\ No newline at end of file
diff --git a/.idea/modules.xml b/.idea/modules.xml
new file mode 100644
index 0000000000000000000000000000000000000000..482c611f35883d7c1998edbfac68f5b5b74300db
--- /dev/null
+++ b/.idea/modules.xml
@@ -0,0 +1,8 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+ <component name="ProjectModuleManager">
+ <modules>
+ <module fileurl="file://$PROJECT_DIR$/.idea/bayesvalidrox.iml" filepath="$PROJECT_DIR$/.idea/bayesvalidrox.iml" />
+ </modules>
+ </component>
+</project>
\ No newline at end of file
diff --git a/.idea/other.xml b/.idea/other.xml
new file mode 100644
index 0000000000000000000000000000000000000000..2e75c2e2af6fe9ab3294dedf305ebb894bb324d7
--- /dev/null
+++ b/.idea/other.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+ <component name="PySciProjectComponent">
+ <option name="PY_INTERACTIVE_PLOTS_SUGGESTED" value="true" />
+ </component>
+</project>
\ No newline at end of file
diff --git a/.idea/vcs.xml b/.idea/vcs.xml
new file mode 100644
index 0000000000000000000000000000000000000000..35eb1ddfbbc029bcab630581847471d7f238ec53
--- /dev/null
+++ b/.idea/vcs.xml
@@ -0,0 +1,6 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<project version="4">
+ <component name="VcsDirectoryMappings">
+ <mapping directory="" vcs="Git" />
+ </component>
+</project>
\ No newline at end of file
diff --git a/docs/diagrams/.$Structure_BayesInf.drawio.bkp b/docs/diagrams/.$Structure_BayesInf.drawio.bkp
new file mode 100644
index 0000000000000000000000000000000000000000..3ba4ed26362c9fa2ea955f14d7ad3f05a15ab00a
--- /dev/null
+++ b/docs/diagrams/.$Structure_BayesInf.drawio.bkp
@@ -0,0 +1,908 @@
+<mxfile host="Electron" modified="2024-04-19T15:13:43.060Z" agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) draw.io/22.1.11 Chrome/114.0.5735.289 Electron/25.9.8 Safari/537.36" etag="DE_l5njUGrsyMG_jufaX" version="22.1.11" type="device" pages="3">
+ <diagram name="Class and function structure" id="efOe0Jku58RX-i1bv-3b">
+ <mxGraphModel dx="2718" dy="686" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-22" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>MCMC</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1270" y="360" width="770" height="380" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-1" value="_kernel_rbf" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1020" y="200" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-2" value="_logpdf" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="820" y="140" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-10" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>BayesInf</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-120" y="290" width="1310" height="680" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-24" value="if self.bootstrap <br>or self.bayes_loocv <br>or self.just_analysis" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=1;entryY=0.5;entryDx=0;entryDy=0;labelBackgroundColor=#ffae00;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-13">
+ <mxGeometry x="0.2902" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-31" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-42" value="if self.name != 'valid'<br>and self.inference_method != 'rejection'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-31">
+ <mxGeometry x="0.5646" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-32" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="HiMKSJFquRK0mIlwyRFI-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-43" value="if self.inference_method == 'mcmc'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-32">
+ <mxGeometry x="-0.0958" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-19">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-52" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#C2C2C2;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-33">
+ <mxGeometry x="-0.112" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-47" value="if self.plot_post_pred" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-34">
+ <mxGeometry x="0.2399" y="-1" relative="1" as="geometry">
+ <mxPoint y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-35" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-20">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-46" value="if self.plot_map_pred" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-35">
+ <mxGeometry x="0.4183" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-55" value="if self.bootstrap" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-54">
+ <mxGeometry x="0.1816" y="3" relative="1" as="geometry">
+ <mxPoint x="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-57" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-56">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-58" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-57">
+ <mxGeometry x="0.7182" y="2" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-60" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-59">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-61" value="if self.error_model<br>and self.name == 'calib'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-60">
+ <mxGeometry x="0.3024" y="2" relative="1" as="geometry">
+ <mxPoint x="67" y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="HiMKSJFquRK0mIlwyRFI-51">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-55" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-54">
+ <mxGeometry x="0.8253" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-9" value="create_inference" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="405" y="539" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-25" value="if len(self.perturbed_data) == 0" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-14">
+ <mxGeometry x="0.3402" relative="1" as="geometry">
+ <mxPoint y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-27" value="if not self.emulator" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-29" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-16">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-44" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-29">
+ <mxGeometry x="0.4722" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-30" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-41" value="if self.emulator" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-30">
+ <mxGeometry x="0.6143" y="-3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-62" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-59">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="340" y="680" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-63" value="if self.error_model<br>and self.name == 'valid'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-62">
+ <mxGeometry x="-0.3906" relative="1" as="geometry">
+ <mxPoint y="121" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-13" value="perform_bootstrap" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="50" y="335" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-14" value="_perturb_data" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-75" y="460" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-15" value="_eval_model" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1050" y="660" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-16" target="xary-zVek9Bg-A1b1ZmA-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-49" value="if hasattr bias_inputs&nbsp;<br>and not hasattr error_model" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#ffae00;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-38">
+ <mxGeometry x="0.3126" y="-3" relative="1" as="geometry">
+ <mxPoint x="-103" y="31" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-39" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-16" target="xary-zVek9Bg-A1b1ZmA-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-16" value="normpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="650" y="455" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-40" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-17" target="xary-zVek9Bg-A1b1ZmA-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-50" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-40">
+ <mxGeometry x="-0.6073" y="-5" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-17" value="_corr_factor_BME" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="650" y="385" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-18" value="_rejection_sampling" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="280" y="890" width="120" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-26" value="if not self.emulator&nbsp;<br>and not self.inference_method == 'rejection'&nbsp;<br>and self.name == 'calib" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-19" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry x="-0.0559" y="15" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-19" target="xary-zVek9Bg-A1b1ZmA-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-48" value="if sigma2_prior is not None<br>and if hasattr bias_inputs<br>and if not hasattr error_model" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#ffae00;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-37">
+ <mxGeometry x="-0.5544" y="-1" relative="1" as="geometry">
+ <mxPoint x="1" y="-5" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-19" value="_posterior_predictive" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="690" y="589" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-28" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-20" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-45" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-28">
+ <mxGeometry x="0.0517" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-20" value="_plot_max_a_posteriori" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="495" y="790" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-21" value="plot_post_predictive" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="630" y="720" width="120" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-36" value="Note: Arrows indicate function calls, beginning calls the end" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="10" y="10" width="190" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-51" value="Color meanings:<br><span style="white-space: pre;">	</span>red: wrong, change<br><span style="white-space: pre;">	</span>orange: seems off, look at again<br><span style="white-space: pre;">	</span>light beige: has been removed" style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="20" y="70" width="220" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-53" value="plot_log_BME" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="150" y="820" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-56" value="plot_post_params" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="660" y="840" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-59" value="create_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="45" y="740" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-1" value="_check_ranges" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1595" y="280" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-2" value="gelman_rubin" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="250" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-3" value="_iterative_scheme" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="2055" y="620" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-4" target="HiMKSJFquRK0mIlwyRFI-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-24" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-4" target="HiMKSJFquRK0mIlwyRFI-11">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-4" value="_my_ESS" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="100" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-8">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-19" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-10">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-53" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;entryX=1;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-52">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-56" value="if opts_sigma != 'B'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-53">
+ <mxGeometry x="0.7377" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-5" value="run_sampler" style="rounded=0;whiteSpace=wrap;html=1;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="534" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-20" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-6" target="HiMKSJFquRK0mIlwyRFI-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-6" value="log_prior" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1595" y="510" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-15" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-7" target="HiMKSJFquRK0mIlwyRFI-9">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-16" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-15">
+ <mxGeometry x="0.0246" y="2" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-7" value="log_likelihood" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1760" y="539" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-8" target="HiMKSJFquRK0mIlwyRFI-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-17" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-12">
+ <mxGeometry x="0.4587" y="4" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-8" target="HiMKSJFquRK0mIlwyRFI-7">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-18" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-13">
+ <mxGeometry x="0.6826" y="4" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-8" value="log_posterior" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1480" y="610" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-9" value="eval_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1760" y="400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-10" value="train_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1450" y="420" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-23" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-11" target="HiMKSJFquRK0mIlwyRFI-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-11" value="marginal_llk_emcee" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1870" y="620" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-25" value="Never used!" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1880" y="680" width="100" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-26" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>BayesModelComp</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1060" y="380" width="840" height="420" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HC1H8j6nMwEtLoyIrXXk-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.75;entryDx=0;entryDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HC1H8j6nMwEtLoyIrXXk-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HiMKSJFquRK0mIlwyRFI-31">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-27" value="model_comparison_all" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-860" y="566" width="160" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-42" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="xary-zVek9Bg-A1b1ZmA-9">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="-630" y="564" as="sourcePoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-47" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-42">
+ <mxGeometry x="-0.4883" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-29" target="HiMKSJFquRK0mIlwyRFI-30">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-49" value="if perturbed_data is None" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-37">
+ <mxGeometry x="-0.0507" y="4" relative="1" as="geometry">
+ <mxPoint x="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-29" value="generate_dataset" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-510" y="566" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-30" value="_perturb_data" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-340" y="636" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-6" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HC1H8j6nMwEtLoyIrXXk-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HiMKSJFquRK0mIlwyRFI-33">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HC1H8j6nMwEtLoyIrXXk-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-31" value="cal_model_weight" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-835" y="466" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-32" value="plot_just_analysis" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-835" y="736" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-33" value="plot_model_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-980" y="416" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-34" value="plot_bayes_factor" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-410" y="431" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-51" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>Discrepancy</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="360" y="1039.82" width="200" height="130" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-52" value="get_sample" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="400" y="1079.82" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.25;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HiMKSJFquRK0mIlwyRFI-34">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-20" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HC1H8j6nMwEtLoyIrXXk-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.75;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HiMKSJFquRK0mIlwyRFI-29">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-1" value="calc_bayes_factors" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-630" y="466" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-2" value="calc_model_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1030" y="566" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-4" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HiMKSJFquRK0mIlwyRFI-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HC1H8j6nMwEtLoyIrXXk-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-16" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.75;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="xary-zVek9Bg-A1b1ZmA-9">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-23" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HC1H8j6nMwEtLoyIrXXk-16">
+ <mxGeometry x="-0.5478" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-18" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.25;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HC1H8j6nMwEtLoyIrXXk-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HiMKSJFquRK0mIlwyRFI-29">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-3" value="calc_justifiability_analysis" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-860" y="666" width="160" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-17" value="setup" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-630" y="566" width="110" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="sQf09xvhinkT827TE7Va" name="Function structure Engine">
+ <mxGraphModel dx="1436" dy="968" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-1" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>Engine</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="130" y="140" width="1390" height="690" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-2" value="hellinger_distance" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-3" value="logpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1050" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-4" value="subdomain" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="625" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-5" value="start_engine" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="250" y="680" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-32" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-6" target="JXjM7l_erEiZMkSmYBvl-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-6" value="train_normal" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="170" y="420" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-7" target="JXjM7l_erEiZMkSmYBvl-9">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="335" y="335" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-7" target="JXjM7l_erEiZMkSmYBvl-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-7" value="train_sequential" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="170" y="310" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-8" value="eval_metamodel" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="190" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-7" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-19" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-23">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-20" value="if len(obs_data) != 0" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-19">
+ <mxGeometry x="0.8137" relative="1" as="geometry">
+ <mxPoint x="-57" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.25;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-24">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-22" value="if len(obs_data) != 0" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-21">
+ <mxGeometry x="0.7684" y="3" relative="1" as="geometry">
+ <mxPoint x="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-23" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-25">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-24" value="if expdes.valid_model_runs" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-23">
+ <mxGeometry x="0.606" y="3" relative="1" as="geometry">
+ <mxPoint x="-16" y="3" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-25" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-26">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-26" value="if mc_ref and pce" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-25">
+ <mxGeometry x="0.7094" y="-3" relative="1" as="geometry">
+ <mxPoint x="-31" y="-3" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-9" value="train_seq_design" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="315" y="310" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-12" value="util_VarBasedDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="670" y="648" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-28" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-31" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-13">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-39" value="if method == 'bayesactdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-38">
+ <mxGeometry x="-0.6235" y="2" relative="1" as="geometry">
+ <mxPoint x="289" y="2" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-15" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-12">
+ <mxGeometry x="0.7865" y="4" relative="1" as="geometry">
+ <mxPoint x="-91" y="185" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-13" value="util_BayesianActiveDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1020" y="680" width="150" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-14" target="JXjM7l_erEiZMkSmYBvl-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-14" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-16" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-13">
+ <mxGeometry x="0.197" y="-3" relative="1" as="geometry">
+ <mxPoint x="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-14" value="utilBayesianDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="880" y="730" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-12">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-42" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-14">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-43" value="if method == 'bayesoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-42">
+ <mxGeometry x="0.6143" y="-3" relative="1" as="geometry">
+ <mxPoint x="3" y="29" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-15" value="run_util_func" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="660" y="450" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-36" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-12">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-41" value="if method == 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-36">
+ <mxGeometry x="-0.5992" relative="1" as="geometry">
+ <mxPoint x="-197" y="62" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-44" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-13">
+ <mxGeometry relative="1" as="geometry">
+ <Array as="points">
+ <mxPoint x="965" y="590" />
+ <mxPoint x="1095" y="590" />
+ </Array>
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-27" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-14">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-16" value="dual_annealing" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="910" y="450" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=1;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-17" target="JXjM7l_erEiZMkSmYBvl-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-6" value="if exploit _method is 'bayesoptdesign',<br style="border-color: var(--border-color);">'bayesactdesign' or 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-5">
+ <mxGeometry x="0.1312" y="2" relative="1" as="geometry">
+ <mxPoint x="17" y="-2" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-17" value="tradeoff_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="980" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-30" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-4">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-1" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-15">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="790" y="280.0000000000002" as="sourcePoint" />
+ <mxPoint x="690" y="499.9999999999998" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-2" value="if exploit _method is 'bayesoptdesign',<br>'bayesactdesign' or 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-1">
+ <mxGeometry x="0.1579" relative="1" as="geometry">
+ <mxPoint x="-15" y="49" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-3" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-16">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="680" y="205.05882352941194" as="sourcePoint" />
+ <mxPoint x="805" y="779.9999999999998" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-4" value="if explore_method == 'dual annealing'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-3">
+ <mxGeometry x="-0.6061" relative="1" as="geometry">
+ <mxPoint x="270" y="46" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-20">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-10" value="if exploit_method == 'alphabetic'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-9">
+ <mxGeometry x="0.8144" y="1" relative="1" as="geometry">
+ <mxPoint x="74" y="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-18" value="choose_next_sample" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="610" y="210" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-20" value="util_AlphOptDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="330" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-21" value="_normpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="430" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-29" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-22" target="JXjM7l_erEiZMkSmYBvl-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-22" value="_corr_factor_BME" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1130" y="220" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-23" value="_posteriorPlot" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="440" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-27" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-14" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-11">
+ <mxGeometry x="0.0929" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-17" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-22">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-18" value="commented out?" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-17">
+ <mxGeometry x="-0.1477" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-24" value="_BME_Calculator" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="220" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-25" value="_validError" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="510" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-26" value="_error_Mean_Std" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="580" width="110" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="ME5gyYpVqUByTnAIOcMV" name="Parameter and function interaction">
+ <mxGraphModel dx="2049" dy="1366" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-61" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-1" value="engine" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="160" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-3" value="Discrepancy" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="240" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-71" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-4" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-4" value="emulator" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="320" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-57" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-65" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-5" value="name" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-47" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-6" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-6" value="bootstrap" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="480" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-7" value="req_outputs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="560" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-79" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-8" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-8" value="selected_indices" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="640" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-35" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-55" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-67" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-9" value="prior_samples" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="720" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-36" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-11" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-68" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-11" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-11" value="n_prior_samples" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="800" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-12" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-80" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-12" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-12" value="measured_data" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="880" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-58" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-13" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-13" value="inference_method" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="960" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-14" value="mcmc_params" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1040" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-63" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-15" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-15" value="perturbed_data" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1120" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-45" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-16" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-77" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-16" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-16" value="bayes_loocv" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1200" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-64" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-17" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-17" value="n_bootstrap_itrs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1280" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-18" value="bootstrap_noise" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1360" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-46" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-19" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-78" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-19" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-19" value="just_analysis" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1440" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-20" value="valid_metrics" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1520" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-52" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-21" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-21" value="plot_post_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1600" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-51" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-22" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-22" value="plot_map_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1680" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-23" value="max_a_posteriori" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1760" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-24" value="corner_title_fmt" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1840" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-25" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-25" value="out_dir" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1920" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-50" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-26" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-66" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-26" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-26" value="error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2000" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-56" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-27" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-72" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-27" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-27" value="bias_inputs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2080" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-41" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-28" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-28" value="measurement_error" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2160" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-44" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-29" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-81" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-29" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-29" value="sigma2s" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2240" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-30" value="log_likes" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2320" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-82" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-31" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-31" value="dtype" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-32" value="create_inference" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="400" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-40" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-39" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-39" value="n_tot_measurement" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2480" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-43" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-42" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-42" value="Discrepancy" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2560" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-49" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-48" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-59" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-48" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-48" value="posterior_df" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2640" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-53" value="create_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="560" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-60" value="perform_bootstrap" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="720" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-75" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-69" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-69" value="__mean_pce_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2720" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-76" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-70" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-70" value="_std_pce_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2800" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-74" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-73" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-73" value="__model_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2880" width="120" height="60" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+</mxfile>
diff --git a/docs/diagrams/.$Structure_BayesInf.drawio.dtmp b/docs/diagrams/.$Structure_BayesInf.drawio.dtmp
new file mode 100644
index 0000000000000000000000000000000000000000..14663ecb0086f14c8b80083b9ecf9a1df87e7ad5
--- /dev/null
+++ b/docs/diagrams/.$Structure_BayesInf.drawio.dtmp
@@ -0,0 +1,964 @@
+<mxfile host="Electron" modified="2024-04-19T16:08:46.718Z" agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) draw.io/22.1.11 Chrome/114.0.5735.289 Electron/25.9.8 Safari/537.36" etag="QkifmTMxwBp7UqUSeBiS" version="22.1.11" type="device" pages="4">
+ <diagram name="Class and function structure" id="efOe0Jku58RX-i1bv-3b">
+ <mxGraphModel dx="3735" dy="1372" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-22" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>MCMC</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1270" y="360" width="770" height="380" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-1" value="_kernel_rbf" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1020" y="200" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-2" value="_logpdf" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="820" y="140" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-10" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>BayesInf</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-120" y="290" width="1310" height="680" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-24" value="if self.bootstrap <br>or self.bayes_loocv <br>or self.just_analysis" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=1;entryY=0.5;entryDx=0;entryDy=0;labelBackgroundColor=#ffae00;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-13">
+ <mxGeometry x="0.2902" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-31" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-42" value="if self.name != 'valid'<br>and self.inference_method != 'rejection'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-31">
+ <mxGeometry x="0.5646" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-32" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="HiMKSJFquRK0mIlwyRFI-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-43" value="if self.inference_method == 'mcmc'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-32">
+ <mxGeometry x="-0.0958" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-19">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-52" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#C2C2C2;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-33">
+ <mxGeometry x="-0.112" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-47" value="if self.plot_post_pred" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-34">
+ <mxGeometry x="0.2399" y="-1" relative="1" as="geometry">
+ <mxPoint y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-35" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-20">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-46" value="if self.plot_map_pred" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-35">
+ <mxGeometry x="0.4183" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-55" value="if self.bootstrap" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-54">
+ <mxGeometry x="0.1816" y="3" relative="1" as="geometry">
+ <mxPoint x="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-57" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-56">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-58" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-57">
+ <mxGeometry x="0.7182" y="2" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-60" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-59">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-61" value="if self.error_model<br>and self.name == 'calib'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-60">
+ <mxGeometry x="0.3024" y="2" relative="1" as="geometry">
+ <mxPoint x="67" y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="HiMKSJFquRK0mIlwyRFI-51">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-55" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-54">
+ <mxGeometry x="0.8253" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-9" value="create_inference" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="405" y="539" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-25" value="if len(self.perturbed_data) == 0" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-14">
+ <mxGeometry x="0.3402" relative="1" as="geometry">
+ <mxPoint y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-27" value="if not self.emulator" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-29" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-16">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-44" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-29">
+ <mxGeometry x="0.4722" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-30" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-41" value="if self.emulator" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-30">
+ <mxGeometry x="0.6143" y="-3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-62" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-59">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="340" y="680" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-63" value="if self.error_model<br>and self.name == 'valid'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-62">
+ <mxGeometry x="-0.3906" relative="1" as="geometry">
+ <mxPoint y="121" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-13" value="perform_bootstrap" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="50" y="335" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-14" value="_perturb_data" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-75" y="460" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-15" value="_eval_model" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1050" y="660" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-16" target="xary-zVek9Bg-A1b1ZmA-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-49" value="if hasattr bias_inputs&nbsp;<br>and not hasattr error_model" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#ffae00;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-38">
+ <mxGeometry x="0.3126" y="-3" relative="1" as="geometry">
+ <mxPoint x="-103" y="31" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-39" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-16" target="xary-zVek9Bg-A1b1ZmA-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-16" value="normpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="650" y="455" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-40" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-17" target="xary-zVek9Bg-A1b1ZmA-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-50" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-40">
+ <mxGeometry x="-0.6073" y="-5" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-17" value="_corr_factor_BME" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="650" y="385" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-18" value="_rejection_sampling" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="280" y="890" width="120" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-26" value="if not self.emulator&nbsp;<br>and not self.inference_method == 'rejection'&nbsp;<br>and self.name == 'calib" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-19" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry x="-0.0559" y="15" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-19" target="xary-zVek9Bg-A1b1ZmA-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-48" value="if sigma2_prior is not None<br>and if hasattr bias_inputs<br>and if not hasattr error_model" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#ffae00;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-37">
+ <mxGeometry x="-0.5544" y="-1" relative="1" as="geometry">
+ <mxPoint x="1" y="-5" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-19" value="_posterior_predictive" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="690" y="589" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-28" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-20" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-45" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-28">
+ <mxGeometry x="0.0517" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-20" value="_plot_max_a_posteriori" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="495" y="790" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-21" value="plot_post_predictive" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="630" y="720" width="120" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-36" value="Note: Arrows indicate function calls, beginning calls the end" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="10" y="10" width="190" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-51" value="Color meanings:<br><span style="white-space: pre;">	</span>red: wrong, change<br><span style="white-space: pre;">	</span>orange: seems off, look at again<br><span style="white-space: pre;">	</span>light beige: has been removed" style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="20" y="70" width="220" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-53" value="plot_log_BME" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="150" y="820" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-56" value="plot_post_params" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="660" y="840" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-59" value="create_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="45" y="740" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-1" value="_check_ranges" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1595" y="280" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-2" value="gelman_rubin" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="250" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-3" value="_iterative_scheme" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="2055" y="620" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-4" target="HiMKSJFquRK0mIlwyRFI-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-24" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-4" target="HiMKSJFquRK0mIlwyRFI-11">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-4" value="_my_ESS" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="100" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-8">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-19" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-10">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-53" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;entryX=1;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-52">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-56" value="if opts_sigma != 'B'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-53">
+ <mxGeometry x="0.7377" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-5" value="run_sampler" style="rounded=0;whiteSpace=wrap;html=1;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="534" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-20" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-6" target="HiMKSJFquRK0mIlwyRFI-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-6" value="log_prior" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1595" y="510" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-15" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-7" target="HiMKSJFquRK0mIlwyRFI-9">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-16" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-15">
+ <mxGeometry x="0.0246" y="2" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-7" value="log_likelihood" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1760" y="539" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-8" target="HiMKSJFquRK0mIlwyRFI-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-17" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-12">
+ <mxGeometry x="0.4587" y="4" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-8" target="HiMKSJFquRK0mIlwyRFI-7">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-18" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-13">
+ <mxGeometry x="0.6826" y="4" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-8" value="log_posterior" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1480" y="610" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-9" value="eval_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1760" y="400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-10" value="train_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1450" y="420" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-23" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-11" target="HiMKSJFquRK0mIlwyRFI-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-11" value="marginal_llk_emcee" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1870" y="620" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-25" value="Never used!" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1880" y="680" width="100" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-26" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>BayesModelComp</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1096" y="380" width="840" height="420" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HC1H8j6nMwEtLoyIrXXk-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.75;entryDx=0;entryDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HC1H8j6nMwEtLoyIrXXk-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HiMKSJFquRK0mIlwyRFI-31">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-27" value="model_comparison_all" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-896" y="566" width="160" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-42" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="xary-zVek9Bg-A1b1ZmA-9">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="-630" y="564" as="sourcePoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-47" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-42">
+ <mxGeometry x="-0.4883" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-29" target="HiMKSJFquRK0mIlwyRFI-30">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-49" value="if perturbed_data is None" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-37">
+ <mxGeometry x="-0.0507" y="4" relative="1" as="geometry">
+ <mxPoint x="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-29" value="generate_dataset" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-546" y="566" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-30" value="_perturb_data" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-376" y="636" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-6" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HC1H8j6nMwEtLoyIrXXk-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HiMKSJFquRK0mIlwyRFI-33">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HC1H8j6nMwEtLoyIrXXk-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-31" value="cal_model_weight" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-871" y="466" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-32" value="plot_just_analysis" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-871" y="736" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-33" value="plot_model_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1016" y="416" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-34" value="plot_bayes_factor" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-446" y="431" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-51" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>Discrepancy</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="360" y="1039.82" width="200" height="130" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-52" value="get_sample" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="400" y="1079.82" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.25;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HiMKSJFquRK0mIlwyRFI-34">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-20" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HC1H8j6nMwEtLoyIrXXk-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.75;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HiMKSJFquRK0mIlwyRFI-29">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-1" value="calc_bayes_factors" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-666" y="466" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-2" value="calc_model_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1066" y="566" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-4" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HiMKSJFquRK0mIlwyRFI-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HC1H8j6nMwEtLoyIrXXk-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-16" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.75;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="xary-zVek9Bg-A1b1ZmA-9">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-23" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HC1H8j6nMwEtLoyIrXXk-16">
+ <mxGeometry x="-0.5478" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-18" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.25;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HC1H8j6nMwEtLoyIrXXk-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HiMKSJFquRK0mIlwyRFI-29">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-3" value="calc_justifiability_analysis" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-896" y="666" width="160" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-17" value="setup" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-666" y="566" width="110" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="sQf09xvhinkT827TE7Va" name="Function structure Engine">
+ <mxGraphModel dx="1436" dy="968" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-1" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>Engine</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="130" y="140" width="1390" height="690" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-2" value="hellinger_distance" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-3" value="logpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1050" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-4" value="subdomain" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="625" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-5" value="start_engine" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="250" y="680" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-32" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-6" target="JXjM7l_erEiZMkSmYBvl-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-6" value="train_normal" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="170" y="420" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-7" target="JXjM7l_erEiZMkSmYBvl-9">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="335" y="335" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-7" target="JXjM7l_erEiZMkSmYBvl-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-7" value="train_sequential" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="170" y="310" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-8" value="eval_metamodel" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="190" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-7" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-19" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-23">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-20" value="if len(obs_data) != 0" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-19">
+ <mxGeometry x="0.8137" relative="1" as="geometry">
+ <mxPoint x="-57" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.25;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-24">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-22" value="if len(obs_data) != 0" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-21">
+ <mxGeometry x="0.7684" y="3" relative="1" as="geometry">
+ <mxPoint x="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-23" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-25">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-24" value="if expdes.valid_model_runs" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-23">
+ <mxGeometry x="0.606" y="3" relative="1" as="geometry">
+ <mxPoint x="-16" y="3" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-25" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-26">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-26" value="if mc_ref and pce" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-25">
+ <mxGeometry x="0.7094" y="-3" relative="1" as="geometry">
+ <mxPoint x="-31" y="-3" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-9" value="train_seq_design" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="315" y="310" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-12" value="util_VarBasedDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="670" y="648" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-28" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-31" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-13">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-39" value="if method == 'bayesactdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-38">
+ <mxGeometry x="-0.6235" y="2" relative="1" as="geometry">
+ <mxPoint x="289" y="2" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-15" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-12">
+ <mxGeometry x="0.7865" y="4" relative="1" as="geometry">
+ <mxPoint x="-91" y="185" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-13" value="util_BayesianActiveDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1020" y="680" width="150" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-14" target="JXjM7l_erEiZMkSmYBvl-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-14" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-16" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-13">
+ <mxGeometry x="0.197" y="-3" relative="1" as="geometry">
+ <mxPoint x="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-14" value="utilBayesianDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="880" y="730" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-12">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-42" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-14">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-43" value="if method == 'bayesoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-42">
+ <mxGeometry x="0.6143" y="-3" relative="1" as="geometry">
+ <mxPoint x="3" y="29" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-15" value="run_util_func" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="660" y="450" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-36" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-12">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-41" value="if method == 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-36">
+ <mxGeometry x="-0.5992" relative="1" as="geometry">
+ <mxPoint x="-197" y="62" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-44" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-13">
+ <mxGeometry relative="1" as="geometry">
+ <Array as="points">
+ <mxPoint x="965" y="590" />
+ <mxPoint x="1095" y="590" />
+ </Array>
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-27" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-14">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-16" value="dual_annealing" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="910" y="450" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=1;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-17" target="JXjM7l_erEiZMkSmYBvl-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-6" value="if exploit _method is 'bayesoptdesign',<br style="border-color: var(--border-color);">'bayesactdesign' or 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-5">
+ <mxGeometry x="0.1312" y="2" relative="1" as="geometry">
+ <mxPoint x="17" y="-2" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-17" value="tradeoff_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="980" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-30" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-4">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-1" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-15">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="790" y="280.0000000000002" as="sourcePoint" />
+ <mxPoint x="690" y="499.9999999999998" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-2" value="if exploit _method is 'bayesoptdesign',<br>'bayesactdesign' or 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-1">
+ <mxGeometry x="0.1579" relative="1" as="geometry">
+ <mxPoint x="-15" y="49" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-3" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-16">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="680" y="205.05882352941194" as="sourcePoint" />
+ <mxPoint x="805" y="779.9999999999998" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-4" value="if explore_method == 'dual annealing'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-3">
+ <mxGeometry x="-0.6061" relative="1" as="geometry">
+ <mxPoint x="270" y="46" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-20">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-10" value="if exploit_method == 'alphabetic'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-9">
+ <mxGeometry x="0.8144" y="1" relative="1" as="geometry">
+ <mxPoint x="74" y="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-18" value="choose_next_sample" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="610" y="210" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-20" value="util_AlphOptDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="330" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-21" value="_normpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="430" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-29" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-22" target="JXjM7l_erEiZMkSmYBvl-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-22" value="_corr_factor_BME" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1130" y="220" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-23" value="_posteriorPlot" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="440" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-27" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-14" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-11">
+ <mxGeometry x="0.0929" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-17" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-22">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-18" value="commented out?" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-17">
+ <mxGeometry x="-0.1477" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-24" value="_BME_Calculator" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="220" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-25" value="_validError" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="510" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-26" value="_error_Mean_Std" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="580" width="110" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="ME5gyYpVqUByTnAIOcMV" name="Parameter and function interaction">
+ <mxGraphModel dx="2049" dy="1366" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-61" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-1" value="engine" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="160" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-3" value="Discrepancy" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="240" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-71" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-4" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-4" value="emulator" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="320" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-57" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-65" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-5" value="name" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-47" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-6" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-6" value="bootstrap" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="480" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-7" value="req_outputs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="560" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-79" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-8" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-8" value="selected_indices" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="640" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-35" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-55" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-67" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-9" value="prior_samples" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="720" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-36" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-11" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-68" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-11" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-11" value="n_prior_samples" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="800" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-12" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-80" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-12" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-12" value="measured_data" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="880" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-58" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-13" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-13" value="inference_method" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="960" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-14" value="mcmc_params" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1040" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-63" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-15" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-15" value="perturbed_data" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1120" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-45" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-16" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-77" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-16" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-16" value="bayes_loocv" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1200" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-64" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-17" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-17" value="n_bootstrap_itrs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1280" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-18" value="bootstrap_noise" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1360" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-46" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-19" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-78" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-19" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-19" value="just_analysis" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1440" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-20" value="valid_metrics" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1520" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-52" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-21" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-21" value="plot_post_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1600" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-51" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-22" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-22" value="plot_map_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1680" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-23" value="max_a_posteriori" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1760" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-24" value="corner_title_fmt" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1840" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-25" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-25" value="out_dir" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1920" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-50" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-26" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-66" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-26" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-26" value="error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2000" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-56" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-27" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-72" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-27" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-27" value="bias_inputs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2080" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-41" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-28" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-28" value="measurement_error" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2160" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-44" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-29" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-81" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-29" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-29" value="sigma2s" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2240" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-30" value="log_likes" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2320" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-82" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-31" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-31" value="dtype" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-32" value="create_inference" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="400" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-40" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-39" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-39" value="n_tot_measurement" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2480" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-43" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-42" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-42" value="Discrepancy" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2560" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-49" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-48" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-59" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-48" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-48" value="posterior_df" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2640" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-53" value="create_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="560" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-60" value="perform_bootstrap" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="720" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-75" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-69" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-69" value="__mean_pce_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2720" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-76" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-70" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-70" value="_std_pce_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2800" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-74" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-73" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-73" value="__model_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2880" width="120" height="60" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="QgiNX2WXFOBDsDgzoFY9" name="Folder structure">
+ <mxGraphModel dx="1436" dy="968" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="KLYezTmecfuvBG8KQe-n-1" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="140" y="80" width="750" height="550" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-2" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="110" width="700" height="220" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-3" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="370" width="180" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-4" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="440" width="180" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-5" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="500" width="180" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-6" value="adaptPlot" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="190" y="150" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-7" value="apoly_construction" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="280" y="150" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-8" value="bayes_linear" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="440" y="150" width="90" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-9" value="engine" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="550" y="150" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-11" value="eval_rec_rule" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="640" y="150" width="100" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-12" value="exp_designs" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="760" y="150" width="90" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-13" value="exploration" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="190" y="210" width="80" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-14" value="glexindex" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="290" y="210" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-15" value="input_space" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="380" y="210" width="80" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-16" value="inputs" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="480" y="210" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-17" value="meta_model_engine" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="570" y="210" width="160" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+</mxfile>
diff --git a/docs/diagrams/Structure_BayesInf.drawio b/docs/diagrams/Structure_BayesInf.drawio
new file mode 100644
index 0000000000000000000000000000000000000000..2e28cbcb3f3e16bee1b42a18f2b93c6f73a156f7
--- /dev/null
+++ b/docs/diagrams/Structure_BayesInf.drawio
@@ -0,0 +1,964 @@
+<mxfile host="Electron" modified="2024-04-19T16:08:43.773Z" agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) draw.io/22.1.11 Chrome/114.0.5735.289 Electron/25.9.8 Safari/537.36" etag="2ELAo-FvqOEnLxZhqqO4" version="22.1.11" type="device" pages="4">
+ <diagram name="Class and function structure" id="efOe0Jku58RX-i1bv-3b">
+ <mxGraphModel dx="3735" dy="1372" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-22" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>MCMC</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1270" y="360" width="770" height="380" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-1" value="_kernel_rbf" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1020" y="200" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-2" value="_logpdf" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="820" y="140" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-10" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>BayesInf</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-120" y="290" width="1310" height="680" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-24" value="if self.bootstrap <br>or self.bayes_loocv <br>or self.just_analysis" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=1;entryY=0.5;entryDx=0;entryDy=0;labelBackgroundColor=#ffae00;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-13">
+ <mxGeometry x="0.2902" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-31" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-42" value="if self.name != 'valid'<br>and self.inference_method != 'rejection'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-31">
+ <mxGeometry x="0.5646" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-32" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="HiMKSJFquRK0mIlwyRFI-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-43" value="if self.inference_method == 'mcmc'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-32">
+ <mxGeometry x="-0.0958" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-19">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-52" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#C2C2C2;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-33">
+ <mxGeometry x="-0.112" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-47" value="if self.plot_post_pred" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-34">
+ <mxGeometry x="0.2399" y="-1" relative="1" as="geometry">
+ <mxPoint y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-35" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-20">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-46" value="if self.plot_map_pred" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-35">
+ <mxGeometry x="0.4183" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-55" value="if self.bootstrap" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-54">
+ <mxGeometry x="0.1816" y="3" relative="1" as="geometry">
+ <mxPoint x="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-57" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-56">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-58" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-57">
+ <mxGeometry x="0.7182" y="2" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-60" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="xary-zVek9Bg-A1b1ZmA-59">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-61" value="if self.error_model<br>and self.name == 'calib'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-60">
+ <mxGeometry x="0.3024" y="2" relative="1" as="geometry">
+ <mxPoint x="67" y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-9" target="HiMKSJFquRK0mIlwyRFI-51">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-55" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-54">
+ <mxGeometry x="0.8253" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-9" value="create_inference" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="405" y="539" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-25" value="if len(self.perturbed_data) == 0" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-14">
+ <mxGeometry x="0.3402" relative="1" as="geometry">
+ <mxPoint y="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-27" value="if not self.emulator" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-29" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-16">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-44" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-29">
+ <mxGeometry x="0.4722" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-30" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-41" value="if self.emulator" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-30">
+ <mxGeometry x="0.6143" y="-3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-62" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-13" target="xary-zVek9Bg-A1b1ZmA-59">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="340" y="680" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-63" value="if self.error_model<br>and self.name == 'valid'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-62">
+ <mxGeometry x="-0.3906" relative="1" as="geometry">
+ <mxPoint y="121" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-13" value="perform_bootstrap" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="50" y="335" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-14" value="_perturb_data" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-75" y="460" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-15" value="_eval_model" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1050" y="660" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-16" target="xary-zVek9Bg-A1b1ZmA-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-49" value="if hasattr bias_inputs&nbsp;<br>and not hasattr error_model" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#ffae00;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-38">
+ <mxGeometry x="0.3126" y="-3" relative="1" as="geometry">
+ <mxPoint x="-103" y="31" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-39" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-16" target="xary-zVek9Bg-A1b1ZmA-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-16" value="normpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="650" y="455" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-40" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-17" target="xary-zVek9Bg-A1b1ZmA-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-50" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-40">
+ <mxGeometry x="-0.6073" y="-5" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-17" value="_corr_factor_BME" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="650" y="385" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-18" value="_rejection_sampling" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="280" y="890" width="120" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-26" value="if not self.emulator&nbsp;<br>and not self.inference_method == 'rejection'&nbsp;<br>and self.name == 'calib" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-19" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry x="-0.0559" y="15" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-19" target="xary-zVek9Bg-A1b1ZmA-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-48" value="if sigma2_prior is not None<br>and if hasattr bias_inputs<br>and if not hasattr error_model" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#ffae00;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-37">
+ <mxGeometry x="-0.5544" y="-1" relative="1" as="geometry">
+ <mxPoint x="1" y="-5" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-19" value="_posterior_predictive" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="690" y="589" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-28" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="xary-zVek9Bg-A1b1ZmA-20" target="xary-zVek9Bg-A1b1ZmA-15">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-45" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#cdcbcb;" vertex="1" connectable="0" parent="xary-zVek9Bg-A1b1ZmA-28">
+ <mxGeometry x="0.0517" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-20" value="_plot_max_a_posteriori" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="495" y="790" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-21" value="plot_post_predictive" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="630" y="720" width="120" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-36" value="Note: Arrows indicate function calls, beginning calls the end" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="10" y="10" width="190" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-51" value="Color meanings:<br><span style="white-space: pre;">	</span>red: wrong, change<br><span style="white-space: pre;">	</span>orange: seems off, look at again<br><span style="white-space: pre;">	</span>light beige: has been removed" style="text;html=1;strokeColor=none;fillColor=none;align=left;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="20" y="70" width="220" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-53" value="plot_log_BME" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="150" y="820" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-56" value="plot_post_params" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="660" y="840" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="xary-zVek9Bg-A1b1ZmA-59" value="create_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="45" y="740" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-1" value="_check_ranges" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1595" y="280" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-2" value="gelman_rubin" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="250" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-3" value="_iterative_scheme" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="2055" y="620" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-4" target="HiMKSJFquRK0mIlwyRFI-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-24" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-4" target="HiMKSJFquRK0mIlwyRFI-11">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-4" value="_my_ESS" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="100" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-8">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-19" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-10">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-53" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;entryX=1;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-5" target="HiMKSJFquRK0mIlwyRFI-52">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-56" value="if opts_sigma != 'B'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=#FF9A03;" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-53">
+ <mxGeometry x="0.7377" y="1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-5" value="run_sampler" style="rounded=0;whiteSpace=wrap;html=1;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="1350" y="534" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-20" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-6" target="HiMKSJFquRK0mIlwyRFI-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-6" value="log_prior" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1595" y="510" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-15" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-7" target="HiMKSJFquRK0mIlwyRFI-9">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-16" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-15">
+ <mxGeometry x="0.0246" y="2" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-7" value="log_likelihood" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1760" y="539" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-8" target="HiMKSJFquRK0mIlwyRFI-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-17" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-12">
+ <mxGeometry x="0.4587" y="4" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-8" target="HiMKSJFquRK0mIlwyRFI-7">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-18" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-13">
+ <mxGeometry x="0.6826" y="4" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-8" value="log_posterior" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1480" y="610" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-9" value="eval_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1760" y="400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-10" value="train_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="1450" y="420" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-23" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-11" target="HiMKSJFquRK0mIlwyRFI-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-11" value="marginal_llk_emcee" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f9f7ed;strokeColor=#CCC1AA;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1870" y="620" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-25" value="Never used!" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#CCC1AA;" vertex="1" parent="1">
+ <mxGeometry x="1880" y="680" width="100" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-26" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>BayesModelComp</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1096" y="380" width="840" height="420" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HC1H8j6nMwEtLoyIrXXk-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0;entryY=0.75;entryDx=0;entryDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HC1H8j6nMwEtLoyIrXXk-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-14" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-27" target="HiMKSJFquRK0mIlwyRFI-31">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-27" value="model_comparison_all" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-896" y="566" width="160" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-42" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="xary-zVek9Bg-A1b1ZmA-9">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="-630" y="564" as="sourcePoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-47" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-42">
+ <mxGeometry x="-0.4883" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-29" target="HiMKSJFquRK0mIlwyRFI-30">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-49" value="if perturbed_data is None" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HiMKSJFquRK0mIlwyRFI-37">
+ <mxGeometry x="-0.0507" y="4" relative="1" as="geometry">
+ <mxPoint x="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-29" value="generate_dataset" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-546" y="566" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-30" value="_perturb_data" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-376" y="636" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-6" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HC1H8j6nMwEtLoyIrXXk-1">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HiMKSJFquRK0mIlwyRFI-33">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HiMKSJFquRK0mIlwyRFI-31" target="HC1H8j6nMwEtLoyIrXXk-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-31" value="cal_model_weight" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-871" y="466" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-32" value="plot_just_analysis" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-871" y="736" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-33" value="plot_model_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1016" y="416" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-34" value="plot_bayes_factor" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-446" y="431" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-51" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>Discrepancy</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="360" y="1039.82" width="200" height="130" as="geometry" />
+ </mxCell>
+ <mxCell id="HiMKSJFquRK0mIlwyRFI-52" value="get_sample" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="400" y="1079.82" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.25;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HiMKSJFquRK0mIlwyRFI-34">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-20" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HC1H8j6nMwEtLoyIrXXk-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.75;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-1" target="HiMKSJFquRK0mIlwyRFI-29">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-1" value="calc_bayes_factors" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-666" y="466" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-2" value="calc_model_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-1066" y="566" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-4" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HiMKSJFquRK0mIlwyRFI-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HC1H8j6nMwEtLoyIrXXk-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-16" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.75;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="xary-zVek9Bg-A1b1ZmA-9">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-23" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="HC1H8j6nMwEtLoyIrXXk-16">
+ <mxGeometry x="-0.5478" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-18" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.25;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HC1H8j6nMwEtLoyIrXXk-17">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-22" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="HC1H8j6nMwEtLoyIrXXk-3" target="HiMKSJFquRK0mIlwyRFI-29">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-3" value="calc_justifiability_analysis" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="-896" y="666" width="160" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="HC1H8j6nMwEtLoyIrXXk-17" value="setup" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="-666" y="566" width="110" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="sQf09xvhinkT827TE7Va" name="Function structure Engine">
+ <mxGraphModel dx="1436" dy="968" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-1" value="<p style="margin:0px;margin-top:4px;text-align:center;"><b>Engine</b></p><hr size="1"><div style="height:2px;"></div>" style="verticalAlign=top;align=left;overflow=fill;fontSize=12;fontFamily=Helvetica;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="130" y="140" width="1390" height="690" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-2" value="hellinger_distance" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-3" value="logpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1050" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-4" value="subdomain" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="625" y="50" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-5" value="start_engine" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="250" y="680" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-32" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-6" target="JXjM7l_erEiZMkSmYBvl-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-6" value="train_normal" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="170" y="420" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-10" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-7" target="JXjM7l_erEiZMkSmYBvl-9">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="335" y="335" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-7" target="JXjM7l_erEiZMkSmYBvl-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-7" value="train_sequential" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="170" y="310" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-8" value="eval_metamodel" style="html=1;whiteSpace=wrap;strokeWidth=2;" vertex="1" parent="1">
+ <mxGeometry x="190" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-7" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-19" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-23">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-20" value="if len(obs_data) != 0" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-19">
+ <mxGeometry x="0.8137" relative="1" as="geometry">
+ <mxPoint x="-57" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-21" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;entryX=0.25;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-24">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-22" value="if len(obs_data) != 0" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-21">
+ <mxGeometry x="0.7684" y="3" relative="1" as="geometry">
+ <mxPoint x="1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-23" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-25">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-24" value="if expdes.valid_model_runs" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-23">
+ <mxGeometry x="0.606" y="3" relative="1" as="geometry">
+ <mxPoint x="-16" y="3" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-25" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-9" target="JXjM7l_erEiZMkSmYBvl-26">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-26" value="if mc_ref and pce" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-25">
+ <mxGeometry x="0.7094" y="-3" relative="1" as="geometry">
+ <mxPoint x="-31" y="-3" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-9" value="train_seq_design" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="315" y="310" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-12" value="util_VarBasedDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="670" y="648" width="130" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-28" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-31" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-13">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-39" value="if method == 'bayesactdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];labelBackgroundColor=default;" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-38">
+ <mxGeometry x="-0.6235" y="2" relative="1" as="geometry">
+ <mxPoint x="289" y="2" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-12" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-13" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-15" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-12">
+ <mxGeometry x="0.7865" y="4" relative="1" as="geometry">
+ <mxPoint x="-91" y="185" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-13" value="util_BayesianActiveDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1020" y="680" width="150" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-14" target="JXjM7l_erEiZMkSmYBvl-6">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-13" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-14" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-16" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-13">
+ <mxGeometry x="0.197" y="-3" relative="1" as="geometry">
+ <mxPoint x="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-14" value="utilBayesianDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="880" y="730" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-12">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-42" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-15" target="JXjM7l_erEiZMkSmYBvl-14">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-43" value="if method == 'bayesoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-42">
+ <mxGeometry x="0.6143" y="-3" relative="1" as="geometry">
+ <mxPoint x="3" y="29" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-15" value="run_util_func" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="660" y="450" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-36" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.25;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-12">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-41" value="if method == 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="JXjM7l_erEiZMkSmYBvl-36">
+ <mxGeometry x="-0.5992" relative="1" as="geometry">
+ <mxPoint x="-197" y="62" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-44" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-13">
+ <mxGeometry relative="1" as="geometry">
+ <Array as="points">
+ <mxPoint x="965" y="590" />
+ <mxPoint x="1095" y="590" />
+ </Array>
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-27" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-16" target="JXjM7l_erEiZMkSmYBvl-14">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-16" value="dual_annealing" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="910" y="450" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-5" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;entryX=1;entryY=0.5;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-17" target="JXjM7l_erEiZMkSmYBvl-18">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-6" value="if exploit _method is 'bayesoptdesign',<br style="border-color: var(--border-color);">'bayesactdesign' or 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-5">
+ <mxGeometry x="0.1312" y="2" relative="1" as="geometry">
+ <mxPoint x="17" y="-2" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-17" value="tradeoff_weights" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="980" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-30" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-4">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-1" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-15">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="790" y="280.0000000000002" as="sourcePoint" />
+ <mxPoint x="690" y="499.9999999999998" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-2" value="if exploit _method is 'bayesoptdesign',<br>'bayesactdesign' or 'varoptdesign'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-1">
+ <mxGeometry x="0.1579" relative="1" as="geometry">
+ <mxPoint x="-15" y="49" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-3" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.75;exitY=1;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-16">
+ <mxGeometry relative="1" as="geometry">
+ <mxPoint x="680" y="205.05882352941194" as="sourcePoint" />
+ <mxPoint x="805" y="779.9999999999998" as="targetPoint" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-4" value="if explore_method == 'dual annealing'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-3">
+ <mxGeometry x="-0.6061" relative="1" as="geometry">
+ <mxPoint x="270" y="46" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-9" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-18" target="JXjM7l_erEiZMkSmYBvl-20">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-10" value="if exploit_method == 'alphabetic'" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-9">
+ <mxGeometry x="0.8144" y="1" relative="1" as="geometry">
+ <mxPoint x="74" y="-1" as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-18" value="choose_next_sample" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="610" y="210" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-20" value="util_AlphOptDesign" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="330" y="210" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-21" value="_normpdf" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="430" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-29" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-22" target="JXjM7l_erEiZMkSmYBvl-3">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-22" value="_corr_factor_BME" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1130" y="220" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-23" value="_posteriorPlot" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="440" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-27" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=0;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-2">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-11" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0.5;exitY=1;exitDx=0;exitDy=0;entryX=0.5;entryY=0;entryDx=0;entryDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-21">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-14" value="always" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-11">
+ <mxGeometry x="0.0929" y="-1" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-17" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=0;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="JXjM7l_erEiZMkSmYBvl-24" target="JXjM7l_erEiZMkSmYBvl-22">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="W5_FOelZ0qj-h3Gb0n3K-18" value="commented out?" style="edgeLabel;html=1;align=center;verticalAlign=middle;resizable=0;points=[];" vertex="1" connectable="0" parent="W5_FOelZ0qj-h3Gb0n3K-17">
+ <mxGeometry x="-0.1477" y="3" relative="1" as="geometry">
+ <mxPoint as="offset" />
+ </mxGeometry>
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-24" value="_BME_Calculator" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="1340" y="220" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-25" value="_validError" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="510" width="110" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="JXjM7l_erEiZMkSmYBvl-26" value="_error_Mean_Std" style="html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="520" y="580" width="110" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="ME5gyYpVqUByTnAIOcMV" name="Parameter and function interaction">
+ <mxGraphModel dx="2049" dy="1366" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-33" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-54" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-61" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-1" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-1" value="engine" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="160" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-3" value="Discrepancy" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="240" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-71" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-4" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-4" value="emulator" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="320" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-37" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-57" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-65" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-5" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-5" value="name" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-47" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;entryX=0.5;entryY=1;entryDx=0;entryDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-6" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-6" value="bootstrap" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="480" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-7" value="req_outputs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="560" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-79" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-8" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-8" value="selected_indices" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="640" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-35" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-55" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-67" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-9" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-9" value="prior_samples" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="720" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-36" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-11" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-68" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-11" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-11" value="n_prior_samples" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="800" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-38" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-12" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-80" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-12" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-12" value="measured_data" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="880" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-58" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-13" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-13" value="inference_method" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="960" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-14" value="mcmc_params" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1040" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-63" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-15" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-15" value="perturbed_data" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1120" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-45" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-16" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-77" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-16" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-16" value="bayes_loocv" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1200" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-64" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-17" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-17" value="n_bootstrap_itrs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1280" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-18" value="bootstrap_noise" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1360" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-46" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-19" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-78" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-19" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-19" value="just_analysis" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1440" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-20" value="valid_metrics" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1520" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-52" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-21" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-21" value="plot_post_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1600" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-51" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-22" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-22" value="plot_map_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1680" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-23" value="max_a_posteriori" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1760" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-24" value="corner_title_fmt" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1840" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-34" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-25" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-25" value="out_dir" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="1920" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-50" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-26" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-66" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-26" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-26" value="error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2000" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-56" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-27" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-72" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-27" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-27" value="bias_inputs" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2080" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-41" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-28" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-28" value="measurement_error" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2160" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-44" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-29" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-81" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-29" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-29" value="sigma2s" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2240" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-30" value="log_likes" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2320" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-82" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-31" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-31" value="dtype" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2400" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-32" value="create_inference" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="400" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-40" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-39" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-39" value="n_tot_measurement" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2480" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-43" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-42" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-42" value="Discrepancy" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2560" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-49" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-48" target="K5oJ7VEt7dPmeK6pba1f-32">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-59" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-48" target="K5oJ7VEt7dPmeK6pba1f-53">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-48" value="posterior_df" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2640" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-53" value="create_error_model" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="560" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-60" value="perform_bootstrap" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="720" y="20" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-75" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-69" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-69" value="__mean_pce_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2720" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-76" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-70" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-70" value="_std_pce_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2800" width="120" height="60" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-74" style="edgeStyle=orthogonalEdgeStyle;rounded=0;orthogonalLoop=1;jettySize=auto;html=1;exitX=1;exitY=0.5;exitDx=0;exitDy=0;" edge="1" parent="1" source="K5oJ7VEt7dPmeK6pba1f-73" target="K5oJ7VEt7dPmeK6pba1f-60">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="K5oJ7VEt7dPmeK6pba1f-73" value="__model_prior_pred" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="40" y="2880" width="120" height="60" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+ <diagram id="QgiNX2WXFOBDsDgzoFY9" name="Folder structure">
+ <mxGraphModel dx="1436" dy="968" grid="1" gridSize="10" guides="1" tooltips="1" connect="1" arrows="1" fold="1" page="1" pageScale="1" pageWidth="850" pageHeight="1100" math="0" shadow="0">
+ <root>
+ <mxCell id="0" />
+ <mxCell id="1" parent="0" />
+ <mxCell id="KLYezTmecfuvBG8KQe-n-1" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="140" y="80" width="750" height="550" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-2" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="110" width="700" height="220" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-3" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="370" width="180" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-4" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="440" width="180" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-5" value="" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="170" y="500" width="180" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-6" value="adaptPlot" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="190" y="150" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-7" value="apoly_construction" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="280" y="150" width="140" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-8" value="bayes_linear" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="440" y="150" width="90" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-9" value="engine" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="550" y="150" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-11" value="eval_rec_rule" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="640" y="150" width="100" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-12" value="exp_designs" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="760" y="150" width="90" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-13" value="exploration" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="190" y="210" width="80" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-14" value="glexindex" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="290" y="210" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-15" value="input_space" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="380" y="210" width="80" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-16" value="inputs" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="480" y="210" width="70" height="50" as="geometry" />
+ </mxCell>
+ <mxCell id="KLYezTmecfuvBG8KQe-n-17" value="meta_model_engine" style="shape=folder;fontStyle=1;spacingTop=10;tabWidth=40;tabHeight=14;tabPosition=left;html=1;whiteSpace=wrap;" vertex="1" parent="1">
+ <mxGeometry x="570" y="210" width="160" height="50" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+</mxfile>
diff --git a/examples/.coverage b/examples/.coverage
new file mode 100644
index 0000000000000000000000000000000000000000..254e10e4371d703eefec0f0437f9c0575be3f5ec
Binary files /dev/null and b/examples/.coverage differ
diff --git a/examples/analytical-function/example_analytical_function.py b/examples/analytical-function/example_analytical_function.py
index 52e7731b576c42cdd29a705865f0a0389f812654..37900127ab39018eef4af9dc5fa3f0326e2f4eeb 100644
--- a/examples/analytical-function/example_analytical_function.py
+++ b/examples/analytical-function/example_analytical_function.py
@@ -141,7 +141,7 @@ if __name__ == "__main__":
# One-shot (normal) or Sequential Adaptive (sequential) Design
ExpDesign.method = 'sequential'
- ExpDesign.n_init_samples = 100#3*ndim
+ ExpDesign.n_init_samples = 140#00#3*ndim
# Sampling methods
# 1) random 2) latin_hypercube 3) sobol 4) halton 5) hammersley
@@ -227,7 +227,8 @@ if __name__ == "__main__":
MetaModelOpts.ExpDesign = ExpDesign
engine = Engine(MetaModelOpts, Model, ExpDesign)
engine.start_engine()
- engine.train_sequential()
+ #engine.train_sequential()
+ engine.train_normal()
# Load the objects
# with open(f"PCEModel_{Model.name}.pkl", "rb") as input:
@@ -266,12 +267,12 @@ if __name__ == "__main__":
# BayesOpts.selected_indices = [0, 3, 5, 7, 9]
# BME Bootstrap
- # BayesOpts.bootstrap = True
- # BayesOpts.n_bootstrap_itrs = 500
- # BayesOpts.bootstrap_noise = 100
+ BayesOpts.bootstrap = True
+ BayesOpts.n_bootstrap_itrs = 500
+ BayesOpts.bootstrap_noise = 100
# Bayesian cross validation
- # BayesOpts.bayes_loocv = True
+ BayesOpts.bayes_loocv = True # TODO: test what this does
# Select the inference method
import emcee
@@ -296,31 +297,34 @@ if __name__ == "__main__":
BayesOpts.Discrepancy = DiscrepancyOpts
# -- (Option C) --
- # DiscOutputOpts = Input()
- # # # OutputName = 'Z'
- # DiscOutputOpts.add_marginals()
- # DiscOutputOpts.Marginals[0].Nnme = '$\sigma^2_{\epsilon}$'
- # DiscOutputOpts.Marginals[0].dist_type = 'uniform'
- # DiscOutputOpts.Marginals[0].parameters = [0, 10]
- # BayesOpts.Discrepancy = {'known': DiscrepancyOpts,
- # 'infer': Discrepancy(DiscOutputOpts)}
-
- # BayesOpts.bias_inputs = {'Z':np.arange(0, 10, 1.).reshape(-1,1) / 9}
- # DiscOutputOpts = Input()
- # # OutputName = 'lambda'
- # DiscOutputOpts.add_marginals()
- # DiscOutputOpts.Marginals[0].name = '$\lambda$'
- # DiscOutputOpts.Marginals[0].dist_type = 'uniform'
- # DiscOutputOpts.Marginals[0].parameters = [0, 1]
-
- # # OutputName = 'sigma_f'
- # DiscOutputOpts.add_marginals()
- # DiscOutputOpts.Marginals[1].Name = '$\sigma_f$'
- # DiscOutputOpts.Marginals[1].dist_type = 'uniform'
- # DiscOutputOpts.Marginals[1].parameters = [0, 1e-4]
- # BayesOpts.Discrepancy = Discrepancy(DiscOutputOpts)
- # BayesOpts.Discrepancy = {'known': DiscrepancyOpts,
- # 'infer': Discrepancy(DiscOutputOpts)}
+ if 0:
+ DiscOutputOpts = Input()
+ # # # OutputName = 'Z'
+ DiscOutputOpts.add_marginals()
+ DiscOutputOpts.Marginals[0].Nnme = '$\sigma^2_{\epsilon}$'
+ DiscOutputOpts.Marginals[0].dist_type = 'uniform'
+ DiscOutputOpts.Marginals[0].parameters = [0, 10]
+ #BayesOpts.Discrepancy = {'known': DiscrepancyOpts,
+ # 'infer': Discrepancy(DiscOutputOpts)}
+
+ BayesOpts.bias_inputs = {'Z':np.arange(0, 10, 1.).reshape(-1,1) / 9}
+
+ DiscOutputOpts = Input()
+ # OutputName = 'lambda'
+ DiscOutputOpts.add_marginals()
+ DiscOutputOpts.Marginals[0].name = '$\lambda$'
+ DiscOutputOpts.Marginals[0].dist_type = 'uniform'
+ DiscOutputOpts.Marginals[0].parameters = [0, 1]
+
+ # # OutputName = 'sigma_f'
+ DiscOutputOpts.add_marginals()
+ DiscOutputOpts.Marginals[1].Name = '$\sigma_f$'
+ DiscOutputOpts.Marginals[1].dist_type = 'uniform'
+ DiscOutputOpts.Marginals[1].parameters = [0, 1e-4]
+ #BayesOpts.Discrepancy = Discrepancy(DiscOutputOpts)
+ BayesOpts.Discrepancy = {'known': DiscrepancyOpts,
+ 'infer': Discrepancy(DiscOutputOpts)}
+
# Start the calibration/inference
Bayes_PCE = BayesOpts.create_inference()
diff --git a/examples/model-comparison/example_model_comparison.py b/examples/model-comparison/example_model_comparison.py
index ebd80fea82a3caf3ff204b89da96c59737ba502b..d678898c178e9722445f08bd22e3ec93906e2ac7 100644
--- a/examples/model-comparison/example_model_comparison.py
+++ b/examples/model-comparison/example_model_comparison.py
@@ -23,6 +23,7 @@ import pandas as pd
import joblib
import sys
sys.path.append("../../src/bayesvalidrox/")
+sys.path.append("../../src/")
from bayesvalidrox.pylink.pylink import PyLinkForwardModel
from bayesvalidrox.surrogate_models.inputs import Input
@@ -38,7 +39,6 @@ from bayes_inference.bayes_model_comparison import BayesModelComparison
from bayesvalidrox.surrogate_models.engine import Engine
if __name__ == "__main__":
-
# Read data
sigma = 0.6
data = {
@@ -277,10 +277,10 @@ if __name__ == "__main__":
"cosine": NL4_engine
}
- # BME Bootstrap optuions
+ # BME Bootstrap options
opts_bootstrap = {
"bootstrap": True,
- "n_samples": 10000,
+ "n_samples": 100,#0,#0, # TODO: difference between this and the n_bootstrap set below?
"Discrepancy": DiscrepancyOpts,
"emulator": True,
"plot_post_pred": False
@@ -289,10 +289,10 @@ if __name__ == "__main__":
# Run model comparison
BayesOpts = BayesModelComparison(
justifiability=True,
- n_bootstarp=100,#00,
- just_n_meas=2
+ n_bootstrap=100,#0,#00,
+ #just_n_meas=2
)
- output_dict = BayesOpts.create_model_comparison(
+ output_dict = BayesOpts.model_comparison_all(
meta_models,
opts_bootstrap
)
diff --git a/src/bayesvalidrox.egg-info/SOURCES.txt b/src/bayesvalidrox.egg-info/SOURCES.txt
index d6619704eee21931221fa73b5d2076a2dce99991..344e9840627bb3e5a89593dbd9256472a8ef41d9 100644
--- a/src/bayesvalidrox.egg-info/SOURCES.txt
+++ b/src/bayesvalidrox.egg-info/SOURCES.txt
@@ -29,10 +29,13 @@ src/bayesvalidrox/surrogate_models/exploration.py
src/bayesvalidrox/surrogate_models/glexindex.py
src/bayesvalidrox/surrogate_models/input_space.py
src/bayesvalidrox/surrogate_models/inputs.py
+src/bayesvalidrox/surrogate_models/meta_model_engine.py
src/bayesvalidrox/surrogate_models/orthogonal_matching_pursuit.py
src/bayesvalidrox/surrogate_models/reg_fast_ard.py
src/bayesvalidrox/surrogate_models/reg_fast_laplace.py
+src/bayesvalidrox/surrogate_models/sequential_design.py
src/bayesvalidrox/surrogate_models/surrogate_models.py
+tests/test_BayesModelComparison.py
tests/test_Discrepancy.py
tests/test_ExpDesign.py
tests/test_Input.py
diff --git a/src/bayesvalidrox/bayes_inference/bayes_inference.py b/src/bayesvalidrox/bayes_inference/bayes_inference.py
index 1898a8ae619597d92bc355ac4249f57019f0aed7..c7cfe78cd08431755f0d16a093363380801964ab 100644
--- a/src/bayesvalidrox/bayes_inference/bayes_inference.py
+++ b/src/bayesvalidrox/bayes_inference/bayes_inference.py
@@ -1,25 +1,25 @@
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
-import numpy as np
-import os
import copy
-import pandas as pd
-from tqdm import tqdm
-from scipy import stats
-import scipy.linalg as spla
-import joblib
-import seaborn as sns
+import gc
+import multiprocessing
+import os
+
import corner
import h5py
-import multiprocessing
-import gc
-from sklearn.metrics import mean_squared_error, r2_score
-from sklearn import preprocessing
-from matplotlib.patches import Patch
import matplotlib.lines as mlines
-from matplotlib.backends.backend_pdf import PdfPages
import matplotlib.pylab as plt
+import numpy as np
+import pandas as pd
+import scipy.linalg as spla
+import seaborn as sns
+from matplotlib.backends.backend_pdf import PdfPages
+from matplotlib.patches import Patch
+from scipy import stats
+from sklearn import preprocessing
+from sklearn.metrics import mean_squared_error, r2_score
+from tqdm import tqdm
from .mcmc import MCMC
@@ -28,6 +28,92 @@ plt.style.use(os.path.join(os.path.split(__file__)[0],
'../', 'bayesvalidrox.mplstyle'))
+# -------------------------------------------------------------------------
+def _kernel_rbf(X, hyperparameters):
+ """
+ Isotropic squared exponential kernel.
+
+ Higher l values lead to smoother functions and therefore to coarser
+ approximations of the training data. Lower l values make functions
+ more wiggly with wide uncertainty regions between training data points.
+
+ sigma_f controls the marginal variance of b(x)
+
+ Parameters
+ ----------
+ X : ndarray of shape (n_samples_X, n_features)
+
+ hyperparameters : Dict
+ Lambda characteristic length
+ sigma_f controls the marginal variance of b(x)
+ sigma_0 unresolvable error nugget term, interpreted as random
+ error that cannot be attributed to measurement error.
+ Returns
+ -------
+ var_cov_matrix : ndarray of shape (n_samples_X,n_samples_X)
+ Kernel k(X, X).
+
+ """
+ from sklearn.gaussian_process.kernels import RBF
+ min_max_scaler = preprocessing.MinMaxScaler()
+ X_minmax = min_max_scaler.fit_transform(X)
+
+ nparams = len(hyperparameters)
+ if nparams < 3:
+ raise AttributeError('Provide 3 parameters for the RBF kernel!')
+
+ # characteristic length (0,1]
+ Lambda = hyperparameters[0]
+ # sigma_f controls the marginal variance of b(x)
+ sigma2_f = hyperparameters[1]
+
+ rbf = RBF(length_scale=Lambda)
+ cov_matrix = sigma2_f * rbf(X_minmax)
+
+ # (unresolvable error) nugget term that is interpreted as random
+ # error that cannot be attributed to measurement error.
+ sigma2_0 = hyperparameters[2:]
+ for i, j in np.ndindex(cov_matrix.shape):
+ cov_matrix[i, j] += np.sum(sigma2_0) if i == j else 0
+
+ return cov_matrix
+
+
+# -------------------------------------------------------------------------
+def _logpdf(x, mean, cov):
+ """
+ Computes the likelihood based on a multivariate normal distribution.
+
+ Parameters
+ ----------
+ x : TYPE
+ DESCRIPTION.
+ mean : array_like
+ Observation data.
+ cov : 2d array
+ Covariance matrix of the distribution.
+
+ Returns
+ -------
+ log_lik : float
+ Log likelihood.
+
+ """
+
+ # Tranform into np arrays
+ x = np.array(x)
+ mean = np.array(mean)
+ cov = np.array(cov)
+
+ n = len(mean)
+ L = spla.cholesky(cov, lower=True)
+ beta = np.sum(np.log(np.diag(L)))
+ dev = x - mean
+ alpha = dev.dot(spla.cho_solve((L, True), dev))
+ log_lik = -0.5 * alpha - beta - n / 2. * np.log(2 * np.pi)
+ return log_lik
+
+
class BayesInference:
"""
A class to perform Bayesian Analysis.
@@ -42,7 +128,7 @@ class BayesInference:
of the variance matrix for a multivariate normal likelihood.
name : str, optional
The type of analysis, either calibration (`Calib`) or validation
- (`Valid`). The default is `'Calib'`.
+ (`Valid`). The default is `'Calib'`. # TODO: what is going on here for validation?
emulator : bool, optional
Analysis with emulator (MetaModel). The default is `True`.
bootstrap : bool, optional
@@ -55,11 +141,11 @@ class BayesInference:
A dictionary with the selected indices of each model output. The
default is `None`. If `None`, all measurement points are used in the
analysis.
- samples : array of shape (n_samples, n_params), optional
+ prior_samples : array of shape (n_samples, n_params), optional
The samples to be used in the analysis. The default is `None`. If
None the samples are drawn from the probablistic input parameter
object of the MetaModel object.
- n_samples : int, optional
+ n_prior_samples : int, optional
Number of samples to be used in the analysis. The default is `500000`.
If samples is not `None`, this argument will be assigned based on the
number of samples given.
@@ -125,26 +211,32 @@ class BayesInference:
"""
- def __init__(self, engine, MetaModel = None, discrepancy=None, emulator=True,
+ def __init__(self, engine, discrepancy=None, emulator=True,
name='Calib', bootstrap=False, req_outputs=None,
- selected_indices=None, samples=None, n_samples=100000,
+ selected_indices=None, prior_samples=None, n_prior_samples=100000,
measured_data=None, inference_method='rejection',
mcmc_params=None, bayes_loocv=False, n_bootstrap_itrs=1,
perturbed_data=[], bootstrap_noise=0.05, just_analysis=False,
valid_metrics=['BME'], plot_post_pred=True,
plot_map_pred=False, max_a_posteriori='mean',
- corner_title_fmt='.2e'):
-
+ corner_title_fmt='.2e', out_dir='', bmc=False):
+
+ self.log_BME_tom = None
+ self.inf_entropy = None
+ self.log_BME = None
+ self.KLD = None
+ self.__mean_pce_prior_pred = None
+ if perturbed_data is None:
+ perturbed_data = []
self.engine = engine
- self.MetaModel = engine.MetaModel
self.Discrepancy = discrepancy
self.emulator = emulator
self.name = name
self.bootstrap = bootstrap
self.req_outputs = req_outputs
self.selected_indices = selected_indices
- self.samples = samples
- self.n_samples = n_samples
+ self.prior_samples = prior_samples
+ self.n_prior_samples = n_prior_samples
self.measured_data = measured_data
self.inference_method = inference_method
self.mcmc_params = mcmc_params
@@ -158,44 +250,63 @@ class BayesInference:
self.plot_map_pred = plot_map_pred
self.max_a_posteriori = max_a_posteriori
self.corner_title_fmt = corner_title_fmt
+ self.out_dir = out_dir
+
+ # Other properties and parameters (found in code, but never set)
+ self.error_model = False # TODO: no example or use case for this!
+ self.bias_inputs = None
+ self.measurement_error = None # TODO: what is this?
+ self.sigma2s = None
+ self.log_likes = None
+ self.n_tot_measurement = None
+ self.Discrepancy = None
+ self.posterior_df = None
+ self.error_MetaModel = None
+ self._mean_pce_prior_pred = None
+ self._std_pce_prior_pred = None
+ self.__model_prior_pred = None
+ self.bmc = bmc # Set to True, if you want to cut short to only Model Comparison
+
+ # System settings
+ if os.name == 'nt':
+ print('')
+ print('WARNING: Performing the inference on windows can lead to reduced accuracy!')
+ print('')
+ self.dtype = np.longdouble
+ else:
+ self.dtype = np.float128
- # -------------------------------------------------------------------------
- def create_inference(self):
+ def setup_inference(self):
"""
- Starts the inference.
-
- Returns
- -------
- BayesInference : obj
- The Bayes inference object.
-
+ This function sets up the inference by checking the inputs and getting
+ needed data.
"""
-
- # Set some variables
- MetaModel = self.MetaModel
Model = self.engine.Model
- n_params = MetaModel.n_params
- output_names = Model.Output.names
- par_names = self.engine.ExpDesign.par_names
- # If the prior is set by the user, take it.
- if self.samples is None:
- self.samples = self.engine.ExpDesign.generate_samples(
- self.n_samples, 'random')
+ # Create output directory
+ if self.out_dir == '':
+ self.out_dir = f'Outputs_Bayes_{self.engine.Model.name}_{self.name}'
+ os.makedirs(self.out_dir, exist_ok=True)
+
+ # If the prior is set by the user, take it, else generate from ExpDes
+ if self.prior_samples is None:
+ self.prior_samples = self.engine.ExpDesign.generate_samples(
+ self.n_prior_samples, 'random')
else:
try:
- samples = self.samples.values
+ samples = self.prior_samples.values
except AttributeError:
- samples = self.samples
+ samples = self.prior_samples
# Take care of an additional Sigma2s
- self.samples = samples[:, :n_params]
+ self.prior_samples = samples[:, :self.engine.MetaModel.n_params]
# Update number of samples
- self.n_samples = self.samples.shape[0]
+ self.n_prior_samples = self.prior_samples.shape[0]
# ---------- Preparation of observation data ----------
- # Read observation data and perturb it if requested.
+ # Read observation data
+ # TODO: later use valid #of measurements. but here only get the model observations?
if self.measured_data is None:
self.measured_data = Model.read_observation(case=self.name)
# Convert measured_data to a data frame
@@ -205,11 +316,13 @@ class BayesInference:
# Extract the total number of measurement points
if self.name.lower() == 'calib':
self.n_tot_measurement = Model.n_obs
- else:
+ elif self.name.lower() == 'valid':
self.n_tot_measurement = Model.n_obs_valid
+ else:
+ raise AttributeError('The set inference type is not known! Use either `calib` or `valid`')
# Find measurement error (if not given) for post predictive plot
- if not hasattr(self, 'measurement_error'):
+ if self.measurement_error is None:
if isinstance(self.Discrepancy, dict):
Disc = self.Discrepancy['known']
else:
@@ -222,383 +335,368 @@ class BayesInference:
self.measurement_error = np.sqrt(Disc.parameters)
except TypeError:
pass
+ # TODO: need a transformation for given measurement error?
- # ---------- Preparation of variance for covariance matrix ----------
- # Independent and identically distributed
- total_sigma2 = dict()
+ # Get Discrepancy type
opt_sigma_flag = isinstance(self.Discrepancy, dict)
opt_sigma = None
- for key_idx, key in enumerate(output_names):
+ # Option A: known error with unknown bias term
+ if opt_sigma_flag and opt_sigma is None:
+ opt_sigma = 'A'
+ # Option B: The sigma2 is known (no bias term)
+ elif self.Discrepancy.parameters is not None:
+ opt_sigma = 'B'
+ # Option C: The sigma2 is unknown (bias term including error)
+ elif not isinstance(self.Discrepancy.InputDisc, str):
+ opt_sigma = 'C'
+ self.Discrepancy.opt_sigma = opt_sigma
+
+ # Set MCMC params if used
+ if self.inference_method.lower() == 'mcmc':
+ if self.mcmc_params is None:
+ self.mcmc_params = {}
+ par_list = ['init_samples', 'n_walkers', 'n_burn', 'n_steps',
+ 'moves', 'multiprocessing', 'verbose']
+ init_val = [None, 100, 200, 100000, None, False, False]
+ for i in range(len(par_list)):
+ if par_list[i] not in list(self.mcmc_params.keys()):
+ self.mcmc_params[par_list[i]] = init_val[i]
+ # -------------------------------------------------------------------------
+ def create_inference(self):
+ """
+ Starts the inference.
+
+ Returns
+ -------
+ BayesInference : obj
+ The Bayes inference object.
+
+ # TODO: should this function really return the class?
+
+ """
+ # Do general set up and check some parameters
+ self.setup_inference()
+
+ # ---------- Preparation of variance for covariance matrix ----------
+ # Independent and identically distributed # TODO: ??
+ total_sigma2 = dict()
+ opt_sigma = self.Discrepancy.opt_sigma
+ for key_idx, key in enumerate(self.engine.Model.Output.names):
# Find opt_sigma
- if opt_sigma_flag and opt_sigma is None:
- # Option A: known error with unknown bias term
- opt_sigma = 'A'
+ sigma2 = None
+ if opt_sigma == 'A':
known_discrepancy = self.Discrepancy['known']
self.Discrepancy = self.Discrepancy['infer']
sigma2 = np.array(known_discrepancy.parameters[key])
- elif opt_sigma == 'A' or self.Discrepancy.parameters is not None:
- # Option B: The sigma2 is known (no bias term)
- if opt_sigma == 'A':
- sigma2 = np.array(known_discrepancy.parameters[key])
- else:
- opt_sigma = 'B'
- sigma2 = np.array(self.Discrepancy.parameters[key])
+ elif opt_sigma == 'B':
+ sigma2 = np.array(self.Discrepancy.parameters[key])
- elif not isinstance(self.Discrepancy.InputDisc, str):
- # Option C: The sigma2 is unknown (bias term including error)
- opt_sigma = 'C'
- self.Discrepancy.opt_sigma = opt_sigma
+ elif opt_sigma == 'C':
n_measurement = self.measured_data[key].values.shape
sigma2 = np.zeros((n_measurement[0]))
-
total_sigma2[key] = sigma2
- self.Discrepancy.opt_sigma = opt_sigma
- self.Discrepancy.total_sigma2 = total_sigma2
+ self.Discrepancy.total_sigma2 = total_sigma2
# If inferred sigma2s obtained from e.g. calibration are given
try:
- self.sigma2s = self.Discrepancy.get_sample(self.n_samples)
+ self.sigma2s = self.Discrepancy.get_sample(self.n_prior_samples)
except:
- pass
+ pass # TODO: should an error be raised in this case? Should this at least be checked against opt_sigma?
# ---------------- Bootstrap & TOM --------------------
if self.bootstrap or self.bayes_loocv or self.just_analysis:
- if len(self.perturbed_data) == 0:
- # zero mean noise Adding some noise to the observation function
- self.perturbed_data = self._perturb_data(
- self.measured_data, output_names
- )
- else:
- self.n_bootstrap_itrs = len(self.perturbed_data)
-
- # -------- Model Discrepancy -----------
- if hasattr(self, 'error_model') and self.error_model \
- and self.name.lower() != 'calib':
- # Select posterior mean as MAP
- MAP_theta = self.samples.mean(axis=0).reshape((1, n_params))
- # MAP_theta = stats.mode(self.samples,axis=0)[0]
-
- # Evaluate the (meta-)model at the MAP
- y_MAP, y_std_MAP = MetaModel.eval_metamodel(samples=MAP_theta)
-
- # Train a GPR meta-model using MAP
- self.error_MetaModel = MetaModel.create_model_error(
- self.bias_inputs, y_MAP, Name=self.name
- )
+ self.perform_bootstrap(total_sigma2)
+ if self.bmc:
+ return self
+ else:
+ print('No bootstrap for TOM performed!') # TODO: stop the code? Use n_bootstrap = 1?
- # -----------------------------------------------------
- # ----- Loop over the perturbed observation data ------
- # -----------------------------------------------------
- # Initilize arrays
- logLikelihoods = np.zeros((self.n_samples, self.n_bootstrap_itrs),
- dtype=np.float16)
- BME_Corr = np.zeros((self.n_bootstrap_itrs))
- log_BME = np.zeros((self.n_bootstrap_itrs))
- KLD = np.zeros((self.n_bootstrap_itrs))
- inf_entropy = np.zeros((self.n_bootstrap_itrs))
-
- # Compute the prior predtions
- # Evaluate the MetaModel
- if self.emulator:
- y_hat, y_std = MetaModel.eval_metamodel(samples=self.samples)
- self.__mean_pce_prior_pred = y_hat
- self._std_pce_prior_pred = y_std
+ # ---------------- Parameter Bayesian inference ----------------
+ # Convert to a dataframe if samples are provided after calibration.
+ MCMC_Obj = None
+ if self.name.lower() == 'valid':
+ self.posterior_df = pd.DataFrame(self.prior_samples, columns=self.engine.ExpDesign.par_names)
+ # Instantiate the MCMC object
+ elif self.inference_method.lower() == 'mcmc':
+ MCMC_Obj = MCMC(self)
+ self.posterior_df = MCMC_Obj.run_sampler(
+ self.measured_data, total_sigma2
+ )
+ # Rejection sampling
+ elif self.inference_method.lower() == 'rejection':
+ self.posterior_df = self._rejection_sampling()
+ else:
+ raise AttributeError('The chosen inference method is not available!')
- # Correct the predictions with Model discrepancy
- if hasattr(self, 'error_model') and self.error_model:
- y_hat_corr, y_std = self.error_MetaModel.eval_model_error(
- self.bias_inputs, self.__mean_pce_prior_pred
- )
- self.__mean_pce_prior_pred = y_hat_corr
- self._std_pce_prior_pred = y_std
+ # Provide posterior's summary
+ print('\n')
+ print('-' * 15 + 'Posterior summary' + '-' * 15)
+ pd.options.display.max_columns = None
+ pd.options.display.max_rows = None
+ print(self.posterior_df.describe())
+ print('-' * 50)
- # Surrogate model's error using RMSE of test data
- if hasattr(MetaModel, 'rmse'):
- surrError = MetaModel.rmse
- else:
- surrError = None
+ # -------- Model Discrepancy -----------
+ if self.error_model and self.name.lower() == 'calib' and MCMC_Obj is not None: # TODO: where is this used
+ # and what does it actually do there?
+ self.create_error_model(opt_sigma=opt_sigma,
+ type_='posterior', sampler=MCMC_Obj)
- else:
- # Evaluate the original model
- self.__model_prior_pred = self._eval_model(
- samples=self.samples, key='PriorPred'
- )
- surrError = None
+ # -------- Posterior predictive -----------
+ self._posterior_predictive()
- # Start the likelihood-BME computations for the perturbed data
- for itr_idx, data in tqdm(
- enumerate(self.perturbed_data),
- total=self.n_bootstrap_itrs,
- desc="Bootstrapping the BME calculations", ascii=True
- ):
+ # ------------------ Visualization --------------------
+ # Posterior parameters
+ self.plot_post_params(opt_sigma)
- # ---------------- Likelihood calculation ----------------
- if self.emulator:
- model_evals = self.__mean_pce_prior_pred
- else:
- model_evals = self.__model_prior_pred
-
- # Leave one out
- if self.bayes_loocv or self.just_analysis:
- self.selected_indices = np.nonzero(data)[0]
-
- # Prepare data dataframe
- nobs = list(self.measured_data.count().values[1:])
- numbers = list(np.cumsum(nobs))
- indices = list(zip([0] + numbers, numbers))
- data_dict = {
- output_names[i]: data[j:k] for i, (j, k) in
- enumerate(indices)
- }
- #print(output_names)
- #print(indices)
- #print(numbers)
- #print(nobs)
- #print(self.measured_data)
- #for i, (j, k) in enumerate(indices):
- # print(i,j,k)
- #print(data)
- #print(data_dict)
- #stop
-
- # Unknown sigma2
- if opt_sigma == 'C' or hasattr(self, 'sigma2s'):
- logLikelihoods[:, itr_idx] = self.normpdf(
- model_evals, data_dict, total_sigma2,
- sigma2=self.sigma2s, std=surrError
- )
- else:
- # known sigma2
- logLikelihoods[:, itr_idx] = self.normpdf(
- model_evals, data_dict, total_sigma2,
- std=surrError
- )
+ # Plot MAP
+ if self.plot_map_pred:
+ self._plot_max_a_posteriori()
- # ---------------- BME Calculations ----------------
- # BME (log)
- log_BME[itr_idx] = np.log(
- np.nanmean(np.exp(logLikelihoods[:, itr_idx],
- dtype=np.longdouble))#float128))
- )
+ # Plot log_BME dist
+ if self.bootstrap:
+ self.plot_log_BME()
- # BME correction when using Emulator
- if self.emulator:
- BME_Corr[itr_idx] = self.__corr_factor_BME(
- data_dict, total_sigma2, log_BME[itr_idx]
- )
+ # Plot posterior predictive
+ if self.plot_post_pred:
+ self._plot_post_predictive()
- # Rejection Step
- if 'kld' in list(map(str.lower, self.valid_metrics)) and\
- 'inf_entropy' in list(map(str.lower, self.valid_metrics)):
- # Random numbers between 0 and 1
- unif = np.random.rand(1, self.n_samples)[0]
-
- # Reject the poorly performed prior
- Likelihoods = np.exp(logLikelihoods[:, itr_idx],
- dtype=np.float64)
- accepted = (Likelihoods/np.max(Likelihoods)) >= unif
- posterior = self.samples[accepted]
-
- # Posterior-based expectation of likelihoods
- postExpLikelihoods = np.mean(
- logLikelihoods[:, itr_idx][accepted]
- )
+ return self
- # Calculate Kullback-Leibler Divergence
- KLD[itr_idx] = postExpLikelihoods - log_BME[itr_idx]
-
- # Posterior-based expectation of prior densities
- if 'inf_entropy' in list(map(str.lower, self.valid_metrics)):
- n_thread = int(0.875 * multiprocessing.cpu_count())
- with multiprocessing.Pool(n_thread) as p:
- postExpPrior = np.mean(np.concatenate(
- p.map(
- self.engine.ExpDesign.JDist.pdf,
- np.array_split(posterior.T, n_thread, axis=1))
- )
- )
- # Information Entropy based on Entropy paper Eq. 38
- inf_entropy[itr_idx] = log_BME[itr_idx] - postExpPrior - \
- postExpLikelihoods
-
- # Clear memory
- gc.collect(generation=2)
-
- # ---------- Store metrics for perturbed data set ----------------
- # Likelihoods (Size: n_samples, n_bootstrap_itr)
- self.log_likes = logLikelihoods
-
- # BME (log), KLD, infEntropy (Size: 1,n_bootstrap_itr)
- self.log_BME = log_BME
-
- # BMECorrFactor (log) (Size: 1,n_bootstrap_itr)
- if self.emulator:
- self.log_BME_corr_factor = BME_Corr
+ def create_error_model(self, type_='posterior', opt_sigma='B', sampler=None):
+ """
+ Creates an error model in the engine.MetaModel based on input dist
+ samples of the chosen type
- if 'kld' in list(map(str.lower, self.valid_metrics)):
- self.KLD = KLD
- if 'inf_entropy' in list(map(str.lower, self.valid_metrics)):
- self.inf_entropy = inf_entropy
+ Parameters
+ ----------
+ opt_sigma : string, optional
+ Type of uncertainty description, only used if type_=='posterior'.
+ The default is 'B'
+ type_ : string
+ Type of parameter samples to use, either 'prior' or 'posterior'.
+ The default is 'posterior'.
+ sampler : MCMC, optional
+ Should be an MCMC object if type=='posterior' and MCMC is used in
+ the inference.In al other cases this parameter is not needed.
- # BME = BME + BMECorrFactor
- if self.emulator:
- self.log_BME += self.log_BME_corr_factor
+ Returns
+ -------
+ None.
- # ---------------- Parameter Bayesian inference ----------------
- if self.inference_method.lower() == 'mcmc':
- # Instantiate the MCMC object
- MCMC_Obj = MCMC(self)
- self.posterior_df = MCMC_Obj.run_sampler(
- self.measured_data, total_sigma2
- )
+ """
+ n_params = self.engine.MetaModel.n_params
- elif self.name.lower() == 'valid':
- # Convert to a dataframe if samples are provided after calibration.
- self.posterior_df = pd.DataFrame(self.samples, columns=par_names)
+ # Get MAP estimate from prior samples
+ if type_ == 'prior':
+ # Select prior ? mean as MAP
+ MAP_theta = self.prior_samples.mean(axis=0).reshape((1, n_params))
- else:
- # Rejection sampling
- self.posterior_df = self._rejection_sampling()
+ # Evaluate the (meta-)model at the MAP
+ y_MAP, y_std_MAP = self.engine.MetaModel.eval_metamodel(samples=MAP_theta)
- # Provide posterior's summary
- print('\n')
- print('-'*15 + 'Posterior summary' + '-'*15)
- pd.options.display.max_columns = None
- pd.options.display.max_rows = None
- print(self.posterior_df.describe())
- print('-'*50)
+ # Train a GPR meta-model using MAP
+ self.error_MetaModel = self.engine.MetaModel.create_model_error(
+ self.bias_inputs, y_MAP, self.measured_data, name=self.name
+ )
- # -------- Model Discrepancy -----------
- if hasattr(self, 'error_model') and self.error_model \
- and self.name.lower() == 'calib':
+ # Get MAP estimate from posterior samples
+ if type_ == 'posterior':
if self.inference_method.lower() == 'mcmc':
- self.error_MetaModel = MCMC_Obj.error_MetaModel
+ self.error_MetaModel = sampler.error_MetaModel
else:
# Select posterior mean as MAP
if opt_sigma == "B":
posterior_df = self.posterior_df.values
else:
- posterior_df = self.posterior_df.values[:, :-Model.n_outputs]
+ posterior_df = self.posterior_df.values[:, :-self.engine.Model.n_outputs]
# Select posterior mean as Maximum a posteriori
map_theta = posterior_df.mean(axis=0).reshape((1, n_params))
# map_theta = stats.mode(Posterior_df,axis=0)[0]
# Evaluate the (meta-)model at the MAP
- y_MAP, y_std_MAP = MetaModel.eval_metamodel(samples=map_theta)
+ y_MAP, y_std_MAP = self.engine.MetaModel.eval_metamodel(samples=map_theta)
# Train a GPR meta-model using MAP
- self.error_MetaModel = MetaModel.create_model_error(
- self.bias_inputs, y_MAP, Name=self.name
- )
-
- # -------- Posterior perdictive -----------
- self._posterior_predictive()
-
- # -----------------------------------------------------
- # ------------------ Visualization --------------------
- # -----------------------------------------------------
- # Create Output directory, if it doesn't exist already.
- out_dir = f'Outputs_Bayes_{Model.name}_{self.name}'
- os.makedirs(out_dir, exist_ok=True)
-
- # -------- Posteior parameters --------
- if opt_sigma != "B":
- par_names.extend(
- [self.Discrepancy.InputDisc.Marginals[i].name for i
- in range(len(self.Discrepancy.InputDisc.Marginals))]
+ self.error_MetaModel = self.engine.MetaModel.create_model_error(
+ self.bias_inputs, y_MAP, self.measured_data, name=self.name
)
- # Pot with corner
- figPosterior = corner.corner(self.posterior_df.to_numpy(),
- labels=par_names,
- quantiles=[0.15, 0.5, 0.85],
- show_titles=True,
- title_fmt=self.corner_title_fmt,
- labelpad=0.2,
- use_math_text=True,
- title_kwargs={"fontsize": 28},
- plot_datapoints=False,
- plot_density=False,
- fill_contours=True,
- smooth=0.5,
- smooth1d=0.5)
- # Loop over axes and set x limits
- if opt_sigma == "B":
- axes = np.array(figPosterior.axes).reshape(
- (len(par_names), len(par_names))
- )
- for yi in range(len(par_names)):
- ax = axes[yi, yi]
- ax.set_xlim(self.engine.ExpDesign.bound_tuples[yi])
- for xi in range(yi):
- ax = axes[yi, xi]
- ax.set_xlim(self.engine.ExpDesign.bound_tuples[xi])
- plt.close()
+ def perform_bootstrap(self, total_sigma2):
+ """
+ Perform bootstrap to get TOM (??)
+
+ Parameters
+ ----------
+ total_sigma2 : dict
+ Dictionary containing the sigma2 for the training(?) data
+ Returns
+ -------
+ None.
- # Turn off gridlines
- for ax in figPosterior.axes:
- ax.grid(False)
+ """
+ MetaModel = self.engine.MetaModel
+ output_names = self.engine.Model.Output.names
+ opt_sigma = self.Discrepancy.opt_sigma
- if self.emulator:
- plotname = f'/Posterior_Dist_{Model.name}_emulator'
+ # Adding some zero mean noise to the observation function
+ if len(self.perturbed_data) == 0:
+ self.perturbed_data = self._perturb_data(
+ self.measured_data, output_names
+ )
else:
- plotname = f'/Posterior_Dist_{Model.name}'
+ self.n_bootstrap_itrs = len(self.perturbed_data)
- figPosterior.set_size_inches((24, 16))
- figPosterior.savefig(f'./{out_dir}{plotname}.pdf',
- bbox_inches='tight')
+ # -------- Model Discrepancy -----------
+ if self.error_model and self.name.lower() == 'valid': # TODO: what should be set so that this is tested?
+ self.create_error_model(type_='prior')
+ # -----------------------------------------------------
+ # ----- Loop over the perturbed observation data ------
+ # -----------------------------------------------------
+ # Initilize arrays
+ logLikelihoods = np.zeros((self.n_prior_samples, self.n_bootstrap_itrs),
+ dtype=np.float16)
+ BME_Corr = np.zeros(self.n_bootstrap_itrs)
+ log_BME = np.zeros(self.n_bootstrap_itrs)
+ KLD = np.zeros(self.n_bootstrap_itrs)
+ inf_entropy = np.zeros(self.n_bootstrap_itrs)
+
+ # Compute the prior predictions
+ # Evaluate the MetaModel
+ if self.emulator:
+ y_hat, y_std = MetaModel.eval_metamodel(samples=self.prior_samples)
+ self.__mean_pce_prior_pred = y_hat
+ self._std_pce_prior_pred = y_std
+
+ # Correct the predictions with Model discrepancy
+ if self.error_model: # TODO this does not check for calib?
+ y_hat_corr, y_std = self.error_MetaModel.eval_model_error(
+ self.bias_inputs, self.__mean_pce_prior_pred)
+ self.__mean_pce_prior_pred = y_hat_corr
+ self._std_pce_prior_pred = y_std
- # -------- Plot MAP --------
- if self.plot_map_pred:
- self._plot_max_a_posteriori()
+ # Surrogate model's error using RMSE of test data
+ if MetaModel.rmse is not None:
+ surrError = MetaModel.rmse
+ else:
+ surrError = None
+ model_evals = self.__mean_pce_prior_pred
- # -------- Plot log_BME dist --------
- if self.bootstrap:
+ # Evaluate the model
+ else:
+ self.__model_prior_pred = self._eval_model(
+ samples=self.prior_samples, key='PriorPred')
+ model_evals = self.__model_prior_pred
+ surrError = None
+
+ # Start the likelihood-BME computations for the perturbed data
+ for itr_idx, data in tqdm(
+ enumerate(self.perturbed_data),
+ total=self.n_bootstrap_itrs,
+ desc="Bootstrapping the BME calculations", ascii=True
+ ):
+
+ # ---------------- Likelihood calculation ----------------
+ # Leave one out (see form of perturbed data)
+ if self.bayes_loocv or self.just_analysis:
+ # Consider only non-zero entries
+ self.selected_indices = np.nonzero(data)[0]
+
+ # Prepare data dataframe # TODO: what's with this transformation?
+ nobs = list(self.measured_data.count().values[1:])
+ numbers = list(np.cumsum(nobs))
+ indices = list(zip([0] + numbers, numbers))
+ data_dict = {
+ output_names[i]: data[j:k] for i, (j, k) in
+ enumerate(indices)
+ }
+
+ # Unknown sigma2
+ if opt_sigma == 'C' or self.sigma2s is not None:
+ logLikelihoods[:, itr_idx] = self.normpdf(
+ model_evals, data_dict, total_sigma2,
+ sigma2=self.sigma2s, std=surrError
+ )
+ else:
+ # known sigma2
+ logLikelihoods[:, itr_idx] = self.normpdf(
+ model_evals, data_dict, total_sigma2,
+ std=surrError
+ )
+ # ---------------- BME Calculations ----------------
+ # BME (log)
+ log_BME[itr_idx] = np.log(
+ np.nanmean(np.exp(logLikelihoods[:, itr_idx],
+ dtype=self.dtype))
+ )
+
+ # BME correction when using Emulator
+ if self.emulator:
+ BME_Corr[itr_idx] = self._corr_factor_BME(
+ data_dict, total_sigma2, log_BME[itr_idx]
+ )
- # Computing the TOM performance
- self.log_BME_tom = stats.chi2.rvs(
- self.n_tot_measurement, size=self.log_BME.shape[0]
+ # Rejection Step
+ if 'kld' in list(map(str.lower, self.valid_metrics)) and \
+ 'inf_entropy' in list(map(str.lower, self.valid_metrics)): # TODO: why and and not or?
+ # Random numbers between 0 and 1
+ unif = np.random.rand(1, self.n_prior_samples)[0]
+
+ # Reject the poorly performed prior
+ Likelihoods = np.exp(logLikelihoods[:, itr_idx],
+ dtype=np.float64)
+ accepted = (Likelihoods / np.max(Likelihoods)) >= unif
+ posterior = self.prior_samples[accepted]
+
+ # Posterior-based expectation of likelihoods
+ postExpLikelihoods = np.mean(
+ logLikelihoods[:, itr_idx][accepted]
)
- fig, ax = plt.subplots()
- sns.kdeplot(self.log_BME_tom, ax=ax, color="green", shade=True)
- sns.kdeplot(
- self.log_BME, ax=ax, color="blue", shade=True,
- label='Model BME')
+ # Calculate Kullback-Leibler Divergence
+ KLD[itr_idx] = postExpLikelihoods - log_BME[itr_idx]
- ax.set_xlabel('log$_{10}$(BME)')
- ax.set_ylabel('Probability density')
+ # Posterior-based expectation of prior densities
+ if 'inf_entropy' in list(map(str.lower, self.valid_metrics)):
+ n_thread = int(0.875 * multiprocessing.cpu_count())
+ with multiprocessing.Pool(n_thread) as p:
+ postExpPrior = np.mean(np.concatenate(
+ p.map(
+ self.engine.ExpDesign.JDist.pdf,
+ np.array_split(posterior.T, n_thread, axis=1))
+ )
+ )
+ # Information Entropy based on Entropy paper Eq. 38
+ inf_entropy[itr_idx] = log_BME[itr_idx] - postExpPrior - postExpLikelihoods
- legend_elements = [
- Patch(facecolor='green', edgecolor='green', label='TOM BME'),
- Patch(facecolor='blue', edgecolor='blue', label='Model BME')
- ]
- ax.legend(handles=legend_elements)
+ # Clear memory
+ gc.collect(generation=2)
- if self.emulator:
- plotname = f'/BME_hist_{Model.name}_emulator'
- else:
- plotname = f'/BME_hist_{Model.name}'
+ # ---------- Store metrics for perturbed data set ----------------
+ # Likelihoods (Size: n_samples, n_bootstrap_itr)
+ self.log_likes = logLikelihoods
- plt.savefig(f'./{out_dir}{plotname}.pdf', bbox_inches='tight')
- plt.show()
- plt.close()
+ # BME (log), KLD, infEntropy (Size: 1,n_bootstrap_itr)
+ self.log_BME = log_BME
- # -------- Posteior perdictives --------
- if self.plot_post_pred:
- # Plot the posterior predictive
- self._plot_post_predictive()
+ # BMECorrFactor (log) (Size: 1,n_bootstrap_itr)
+ # BME = BME + BMECorrFactor
+ if self.emulator:
+ self.log_BME += BME_Corr
- return self
+ if 'kld' in list(map(str.lower, self.valid_metrics)):
+ self.KLD = KLD
+ if 'inf_entropy' in list(map(str.lower, self.valid_metrics)):
+ self.inf_entropy = inf_entropy
# -------------------------------------------------------------------------
def _perturb_data(self, data, output_names):
"""
- Returns an array with n_bootstrap_itrs rowsof perturbed data.
+ Returns an array with n_bootstrap_itrs rows of perturbed data.
The first row includes the original observation data.
If `self.bayes_loocv` is True, a 2d-array will be returned with
repeated rows and zero diagonal entries.
@@ -608,7 +706,7 @@ class BayesInference:
data : pandas DataFrame
Observation data.
output_names : list
- List of the output names.
+ The output names.
Returns
-------
@@ -620,12 +718,11 @@ class BayesInference:
obs_data = data[output_names].values
n_measurement, n_outs = obs_data.shape
self.n_tot_measurement = obs_data[~np.isnan(obs_data)].shape[0]
- # Number of bootstrap iterations
- if self.bayes_loocv:
- self.n_bootstrap_itrs = self.n_tot_measurement
# Pass loocv dataset
if self.bayes_loocv:
+ # Number of bootstrap iterations
+ self.n_bootstrap_itrs = self.n_tot_measurement
obs = obs_data.T[~np.isnan(obs_data.T)]
final_data = np.repeat(np.atleast_2d(obs), self.n_bootstrap_itrs,
axis=0)
@@ -633,15 +730,18 @@ class BayesInference:
return final_data
else:
+ # Init return data with original data
final_data = np.zeros(
(self.n_bootstrap_itrs, self.n_tot_measurement)
- )
+ )
final_data[0] = obs_data.T[~np.isnan(obs_data.T)]
for itrIdx in range(1, self.n_bootstrap_itrs):
data = np.zeros((n_measurement, n_outs))
for idx in range(len(output_names)):
+ # Perturb the data
std = np.nanstd(obs_data[:, idx])
if std == 0:
+ print('Note: Use std=0.01 for perturbation')
std = 0.001
noise = std * noise_level
data[:, idx] = np.add(
@@ -653,45 +753,17 @@ class BayesInference:
return final_data
- # -------------------------------------------------------------------------
- def _logpdf(self, x, mean, cov):
- """
- computes the likelihood based on a multivariate normal distribution.
-
- Parameters
- ----------
- x : TYPE
- DESCRIPTION.
- mean : array_like
- Observation data.
- cov : 2d array
- Covariance matrix of the distribution.
-
- Returns
- -------
- log_lik : float
- Log likelihood.
-
- """
- n = len(mean)
- L = spla.cholesky(cov, lower=True)
- beta = np.sum(np.log(np.diag(L)))
- dev = x - mean
- alpha = dev.dot(spla.cho_solve((L, True), dev))
- log_lik = -0.5 * alpha - beta - n / 2. * np.log(2 * np.pi)
- return log_lik
-
# -------------------------------------------------------------------------
def _eval_model(self, samples=None, key='MAP'):
"""
- Evaluates Forward Model.
+ Evaluates Forward Model and zips the results
Parameters
----------
samples : array of shape (n_samples, n_params), optional
Parameter sets. The default is None.
key : str, optional
- Key string to be passed to the run_model_parallel method.
+ Descriptive key string for the run_model_parallel method.
The default is 'MAP'.
Returns
@@ -700,18 +772,17 @@ class BayesInference:
Model outputs.
"""
- MetaModel = self.MetaModel
Model = self.engine.Model
if samples is None:
- self.samples = self.engine.ExpDesign.generate_samples(
- self.n_samples, 'random')
+ self.prior_samples = self.engine.ExpDesign.generate_samples(
+ self.n_prior_samples, 'random')
else:
- self.samples = samples
- self.n_samples = len(samples)
+ self.prior_samples = samples
+ self.n_prior_samples = len(samples)
model_outputs, _ = Model.run_model_parallel(
- self.samples, key_str=key+self.name)
+ self.prior_samples, key_str=key + self.name)
# Clean up
# Zip the subdirectories
@@ -724,55 +795,6 @@ class BayesInference:
return model_outputs
- # -------------------------------------------------------------------------
- def _kernel_rbf(self, X, hyperparameters):
- """
- Isotropic squared exponential kernel.
-
- Higher l values lead to smoother functions and therefore to coarser
- approximations of the training data. Lower l values make functions
- more wiggly with wide uncertainty regions between training data points.
-
- sigma_f controls the marginal variance of b(x)
-
- Parameters
- ----------
- X : ndarray of shape (n_samples_X, n_features)
-
- hyperparameters : Dict
- Lambda characteristic length
- sigma_f controls the marginal variance of b(x)
- sigma_0 unresolvable error nugget term, interpreted as random
- error that cannot be attributed to measurement error.
- Returns
- -------
- var_cov_matrix : ndarray of shape (n_samples_X,n_samples_X)
- Kernel k(X, X).
-
- """
- from sklearn.gaussian_process.kernels import RBF
- min_max_scaler = preprocessing.MinMaxScaler()
- X_minmax = min_max_scaler.fit_transform(X)
-
- nparams = len(hyperparameters)
- # characteristic length (0,1]
- Lambda = hyperparameters[0]
- # sigma_f controls the marginal variance of b(x)
- sigma2_f = hyperparameters[1]
-
- # cov_matrix = sigma2_f*rbf_kernel(X_minmax, gamma = 1/Lambda**2)
-
- rbf = RBF(length_scale=Lambda)
- cov_matrix = sigma2_f * rbf(X_minmax)
- if nparams > 2:
- # (unresolvable error) nugget term that is interpreted as random
- # error that cannot be attributed to measurement error.
- sigma2_0 = hyperparameters[2:]
- for i, j in np.ndindex(cov_matrix.shape):
- cov_matrix[i, j] += np.sum(sigma2_0) if i == j else 0
-
- return cov_matrix
-
# -------------------------------------------------------------------------
def normpdf(self, outputs, obs_data, total_sigma2s, sigma2=None, std=None):
"""
@@ -788,7 +810,7 @@ class BayesInference:
A dictionary/dataframe containing the observation data.
total_sigma2s : dict
A dictionary with known values of the covariance diagonal entries,
- a.k.a sigma^2.
+ a.k.a. sigma^2.
sigma2 : array, optional
An array of the sigma^2 samples, when the covariance diagonal
entries are unknown and are being jointly inferred. The default is
@@ -809,11 +831,11 @@ class BayesInference:
# Extract the requested model outputs for likelihood calulation
if self.req_outputs is None:
- req_outputs = Model.Output.names
+ req_outputs = Model.Output.names # TODO: should this then be saved as self.req_outputs?
else:
req_outputs = list(self.req_outputs)
- # Loop over the outputs
+ # Loop over the output keys
for idx, out in enumerate(req_outputs):
# (Meta)Model Output
@@ -825,26 +847,25 @@ class BayesInference:
except AttributeError:
data = obs_data[out][~np.isnan(obs_data[out])]
- # Prepare sigma2s
+ # Prepare data uncertainty / error estimation (sigma2s)
non_nan_indices = ~np.isnan(total_sigma2s[out])
tot_sigma2s = total_sigma2s[out][non_nan_indices][:nout]
- # Add the std of the PCE is chosen as emulator.
+ # Add the std of the PCE if an emulator is used
if self.emulator:
if std is not None:
- tot_sigma2s += std[out]**2
-
- # Covariance Matrix
- covMatrix = np.diag(tot_sigma2s)
+ tot_sigma2s += std[out] ** 2
# Select the data points to compare
try:
indices = self.selected_indices[out]
except:
indices = list(range(nout))
- covMatrix = np.diag(covMatrix[indices, indices])
- # If sigma2 is not given, use given total_sigma2s
+ # Set up Covariance Matrix
+ covMatrix = np.diag(np.diag(tot_sigma2s)[indices, indices])
+
+ # If sigma2 is not given, use given total_sigma2s and move to next itr
if sigma2 is None:
logLik += stats.multivariate_normal.logpdf(
outputs[out][:, indices], data[indices], covMatrix)
@@ -860,26 +881,24 @@ class BayesInference:
# Covariance Matrix
covMatrix = np.diag(tot_sigma2s)
- if sigma2 is not None:
- # Check the type error term
- if hasattr(self, 'bias_inputs') and \
- not hasattr(self, 'error_model'):
- # Infer a Bias model usig Gaussian Process Regression
- bias_inputs = np.hstack(
- (self.bias_inputs[out],
- tot_outputs[s_idx].reshape(-1, 1)))
-
- params = sigma2[s_idx, idx*3:(idx+1)*3]
- covMatrix = self._kernel_rbf(bias_inputs, params)
- else:
- # Infer equal sigma2s
- try:
- sigma_2 = sigma2[s_idx, idx]
- except TypeError:
- sigma_2 = 0.0
+ # Check the type error term
+ if self.bias_inputs is not None and self.error_model is None:
+ # Infer a Bias model usig Gaussian Process Regression
+ bias_inputs = np.hstack(
+ (self.bias_inputs[out],
+ tot_outputs[s_idx].reshape(-1, 1)))
+
+ params = sigma2[s_idx, idx * 3:(idx + 1) * 3]
+ covMatrix = _kernel_rbf(bias_inputs, params)
+ else:
+ # Infer equal sigma2s
+ try:
+ sigma_2 = sigma2[s_idx, idx]
+ except TypeError:
+ sigma_2 = 0.0
- covMatrix += sigma_2 * np.eye(nout)
- # covMatrix = np.diag(sigma2 * total_sigma2s)
+ covMatrix += sigma_2 * np.eye(nout)
+ # covMatrix = np.diag(sigma2 * total_sigma2s)
# Select the data points to compare
try:
@@ -889,86 +908,45 @@ class BayesInference:
covMatrix = np.diag(covMatrix[indices, indices])
# Compute loglikelihood
- logliks[s_idx] = self._logpdf(
+ logliks[s_idx] = _logpdf(
tot_outputs[s_idx, indices], data[indices], covMatrix
- )
-
+ )
logLik += logliks
return logLik
# -------------------------------------------------------------------------
- def _corr_factor_BME_old(self, Data, total_sigma2s, posterior):
+ def _corr_factor_BME(self, obs_data, total_sigma2s, logBME):
"""
Calculates the correction factor for BMEs.
- """
- MetaModel = self.MetaModel
- OrigModelOutput = self.engine.ExpDesign.Y
- Model = self.engine.Model
-
- # Posterior with guassian-likelihood
- postDist = stats.gaussian_kde(posterior.T)
-
- # Remove NaN
- Data = Data[~np.isnan(Data)]
- total_sigma2s = total_sigma2s[~np.isnan(total_sigma2s)]
-
- # Covariance Matrix
- covMatrix = np.diag(total_sigma2s[:self.n_tot_measurement])
-
- # Extract the requested model outputs for likelihood calulation
- if self.req_outputs is None:
- OutputType = Model.Output.names
- else:
- OutputType = list(self.req_outputs)
-
- # SampleSize = OrigModelOutput[OutputType[0]].shape[0]
-
-
- # Flatten the OutputType for OrigModel
- TotalOutputs = np.concatenate([OrigModelOutput[x] for x in OutputType], 1)
-
- NrofBayesSamples = self.n_samples
- # Evaluate MetaModel on the experimental design
- Samples = self.engine.ExpDesign.X
- OutputRS, stdOutputRS = MetaModel.eval_metamodel(samples=Samples)
-
- # Reset the NrofSamples to NrofBayesSamples
- self.n_samples = NrofBayesSamples
-
- # Flatten the OutputType for MetaModel
- TotalPCEOutputs = np.concatenate([OutputRS[x] for x in OutputRS], 1)
- TotalPCEstdOutputRS= np.concatenate([stdOutputRS[x] for x in stdOutputRS], 1)
+
+ Parameters
+ ----------
+ obs_data : dict
+ A dictionary/dataframe containing the observation data.
+ total_sigma2s : dict
+ A dictionary with known values of the covariance diagonal entries,
+ a.k.a sigma^2.
+ logBME : ??
+ The log_BME obtained from the estimated likelihoods
- logweight = 0
- for i, sample in enumerate(Samples):
- # Compute likelilhood output vs RS
- covMatrix = np.diag(TotalPCEstdOutputRS[i]**2)
- logLik = self._logpdf(TotalOutputs[i], TotalPCEOutputs[i], covMatrix)
- # Compute posterior likelihood of the collocation points
- logpostLik = np.log(postDist.pdf(sample[:, None]))[0]
- if logpostLik != -np.inf:
- logweight += logLik + logpostLik
- return logweight
+ Returns
+ -------
+ np.log(weights) : ??
+ Correction factors # TODO: factors or log of factors?
- # -------------------------------------------------------------------------
- def __corr_factor_BME(self, obs_data, total_sigma2s, logBME):
"""
- Calculates the correction factor for BMEs.
- """
- MetaModel = self.MetaModel
+ # Extract the requested model outputs for likelihood calulation
+ MetaModel = self.engine.MetaModel
samples = self.engine.ExpDesign.X
model_outputs = self.engine.ExpDesign.Y
- Model = self.engine.Model
n_samples = samples.shape[0]
-
- # Extract the requested model outputs for likelihood calulation
- output_names = Model.Output.names
+ output_names = self.engine.Model.Output.names
# Evaluate MetaModel on the experimental design and ValidSet
OutputRS, stdOutputRS = MetaModel.eval_metamodel(samples=samples)
- logLik_data = np.zeros((n_samples))
- logLik_model = np.zeros((n_samples))
+ logLik_data = np.zeros(n_samples)
+ logLik_model = np.zeros(n_samples)
# Loop over the outputs
for idx, out in enumerate(output_names):
@@ -997,7 +975,7 @@ class BayesInference:
y_m_hat = OutputRS[out][i]
# CovMatrix with the surrogate error
- covMatrix = np.eye(len(y_m)) * 1/(2*np.pi)
+ covMatrix = np.eye(len(y_m)) * 1 / (2 * np.pi)
# Select the data points to compare
try:
@@ -1008,20 +986,20 @@ class BayesInference:
covMatrix_data = np.diag(covMatrix_data[indices, indices])
# Compute likelilhood output vs data
- logLik_data[i] += self._logpdf(
+ logLik_data[i] += _logpdf(
y_m_hat[indices], data[indices],
covMatrix_data
- )
+ )
# Compute likelilhood output vs surrogate
- logLik_model[i] += self._logpdf(
+ logLik_model[i] += _logpdf(
y_m_hat[indices], y_m[indices],
covMatrix
- )
+ )
# Weight
logLik_data -= logBME
- weights = np.mean(np.exp(logLik_model+logLik_data))
+ weights = np.mean(np.exp(logLik_model + logLik_data))
return np.log(weights)
@@ -1037,45 +1015,43 @@ class BayesInference:
Posterior samples of the input parameters.
"""
+ if self.prior_samples is None:
+ raise AttributeError('No prior samples available!')
- MetaModel = self.MetaModel
+ if self.log_likes is None:
+ raise AttributeError('No log-likelihoods available!')
+
+ # Get sigmas # TODO: is this data uncertainty?
try:
sigma2_prior = self.Discrepancy.sigma2_prior
except:
sigma2_prior = None
- # Check if the discrepancy is defined as a distribution:
- samples = self.samples
-
+ # Combine samples and sigma2 for the return
+ samples = self.prior_samples
if sigma2_prior is not None:
samples = np.hstack((samples, sigma2_prior))
# Take the first column of Likelihoods (Observation data without noise)
if self.just_analysis or self.bayes_loocv:
- index = self.n_tot_measurement-1
- likelihoods = np.exp(self.log_likes[:, index], dtype=np.longdouble)#np.float128)
+ index = self.n_tot_measurement - 1
else:
- likelihoods = np.exp(self.log_likes[:, 0], dtype=np.longdouble)#np.float128)
+ index = 0
+
+ # Use longdouble on windows, float128 on linux
+ likelihoods = np.exp(self.log_likes[:, index], dtype=self.dtype)
n_samples = len(likelihoods)
- norm_ikelihoods = likelihoods / np.max(likelihoods)
+ norm_likelihoods = likelihoods / np.max(likelihoods)
# Normalize based on min if all Likelihoods are zero
if all(likelihoods == 0.0):
likelihoods = self.log_likes[:, 0]
- norm_ikelihoods = likelihoods / np.min(likelihoods)
+ norm_likelihoods = likelihoods / np.min(likelihoods)
- # Random numbers between 0 and 1
+ # Reject the poorly performed prior compared to a uniform distribution
unif = np.random.rand(1, n_samples)[0]
-
- # Reject the poorly performed prior
- accepted_samples = samples[norm_ikelihoods >= unif]
-
- # Output the Posterior
- par_names = self.engine.ExpDesign.par_names
- if sigma2_prior is not None:
- for name in self.Discrepancy.name:
- par_names.append(name)
+ accepted_samples = samples[norm_likelihoods >= unif]
return pd.DataFrame(accepted_samples, columns=sigma2_prior)
@@ -1097,25 +1073,21 @@ class BayesInference:
"""
- MetaModel = self.MetaModel
+ MetaModel = self.engine.MetaModel
Model = self.engine.Model
- # Make a directory to save the prior/posterior predictive
- out_dir = f'Outputs_Bayes_{Model.name}_{self.name}'
- os.makedirs(out_dir, exist_ok=True)
-
- # Read observation data and perturb it if requested
+ # Read observation data and perturb it if requested # TODO: where is the perturbation?
if self.measured_data is None:
self.measured_data = Model.read_observation(case=self.name)
if not isinstance(self.measured_data, pd.DataFrame):
self.measured_data = pd.DataFrame(self.measured_data)
- # X_values
+ # X_values and prior sigma2
x_values = self.engine.ExpDesign.x_values
-
try:
- sigma2_prior = self.Discrepancy.sigma2_prior
+ sigma2_prior = self.Discrepancy.sigma2_prior # TODO: what is this? Looks to be built for a different
+ # Discrepancy structure
except:
sigma2_prior = None
@@ -1123,45 +1095,47 @@ class BayesInference:
posterior_df = self.posterior_df
# Take care of the sigma2
- if sigma2_prior is not None:
+ sigma2s = None
+ if sigma2_prior is not None: # TODO: why is this the if for this code?
try:
- sigma2s = posterior_df[self.Discrepancy.name].values
+ sigma2s = posterior_df[self.Discrepancy.name].values # TODO: what is Discrepancy.name?
posterior_df = posterior_df.drop(
labels=self.Discrepancy.name, axis=1
- )
+ )
except:
sigma2s = self.sigma2s
# Posterior predictive
if self.emulator:
- if self.inference_method == 'rejection':
- prior_pred = self.__mean_pce_prior_pred
- if self.name.lower() != 'calib':
- post_pred = self.__mean_pce_prior_pred
+ if self.inference_method.lower() == 'rejection': # TODO: combine these two? Why is there no
+ # post_pred_std for rejection sampling?
+ prior_pred = self._mean_pce_prior_pred
+ if self.name.lower() == 'valid':
+ post_pred = self._mean_pce_prior_pred
post_pred_std = self._std_pce_prior_pred
else:
- post_pred, post_pred_std = MetaModel.eval_metamodel(
+ post_pred, post_pred_std = MetaModel.eval_metamodel( # TODO: recheck if this is needed
samples=posterior_df.values
- )
+ )
- else:
- if self.inference_method == 'rejection':
+ else: # TODO: see emulator version
+ if self.inference_method.lower() == 'rejection':
prior_pred = self.__model_prior_pred
- if self.name.lower() != 'calib':
+ if self.name.lower() == 'valid':
post_pred = self.__mean_pce_prior_pred,
post_pred_std = self._std_pce_prior_pred
else:
post_pred = self._eval_model(
samples=posterior_df.values, key='PostPred'
- )
+ )
# Correct the predictions with Model discrepancy
- if hasattr(self, 'error_model') and self.error_model:
+ if self.error_model:
y_hat, y_std = self.error_MetaModel.eval_model_error(
self.bias_inputs, post_pred
- )
+ )
post_pred, post_pred_std = y_hat, y_std
- # Add discrepancy from likelihood Sample to the current posterior runs
+ # Add discrepancy from likelihood samples to the current posterior runs
total_sigma2 = self.Discrepancy.total_sigma2
post_pred_withnoise = copy.deepcopy(post_pred)
for varIdx, var in enumerate(Model.Output.names):
@@ -1173,16 +1147,15 @@ class BayesInference:
tot_sigma2 = clean_sigma2[:len(pred)]
cov = np.diag(tot_sigma2)
- # Check the type error term
+ # Account for additional error terms
if sigma2_prior is not None:
# Inferred sigma2s
- if hasattr(self, 'bias_inputs') and \
- not hasattr(self, 'error_model'):
+ if self.bias_inputs is not None and self.error_model is None:
# TODO: Infer a Bias model usig GPR
bias_inputs = np.hstack((
self.bias_inputs[var], pred.reshape(-1, 1)))
- params = sigma2s[i, varIdx*3:(varIdx+1)*3]
- cov = self._kernel_rbf(bias_inputs, params)
+ params = sigma2s[i, varIdx * 3:(varIdx + 1) * 3]
+ cov = _kernel_rbf(bias_inputs, params)
else:
# Infer equal sigma2s
try:
@@ -1193,25 +1166,25 @@ class BayesInference:
# Convert biasSigma2s to a covMatrix
cov += sigma2 * np.eye(len(pred))
+ # Add predictive metamodel error/uncertainty
if self.emulator:
- if hasattr(MetaModel, 'rmse') and \
- MetaModel.rmse is not None:
+ if MetaModel.rmse is not None:
stdPCE = MetaModel.rmse[var]
else:
stdPCE = post_pred_std[var][i]
# Expected value of variance (Assump: i.i.d stds)
- cov += np.diag(stdPCE**2)
+ cov += np.diag(stdPCE ** 2)
# Sample a multivariate normal distribution with mean of
- # prediction and variance of cov
+ # posterior prediction and variance of cov
post_pred_withnoise[var][i] = np.random.multivariate_normal(
pred, cov, 1
- )
+ )
# ----- Prior Predictive -----
if self.inference_method.lower() == 'rejection':
# Create hdf5 metadata
- hdf5file = f'{out_dir}/priorPredictive.hdf5'
+ hdf5file = f'{self.out_dir}/priorPredictive.hdf5'
hdf5_exist = os.path.exists(hdf5file)
if hdf5_exist:
os.remove(hdf5file)
@@ -1232,7 +1205,7 @@ class BayesInference:
# ----- Posterior Predictive only model evaluations -----
# Create hdf5 metadata
- hdf5file = out_dir+'/postPredictive_wo_noise.hdf5'
+ hdf5file = self.out_dir + '/postPredictive_wo_noise.hdf5'
hdf5_exist = os.path.exists(hdf5file)
if hdf5_exist:
os.remove(hdf5file)
@@ -1253,7 +1226,7 @@ class BayesInference:
# ----- Posterior Predictive with noise -----
# Create hdf5 metadata
- hdf5file = out_dir+'/postPredictive.hdf5'
+ hdf5file = self.out_dir + '/postPredictive.hdf5'
hdf5_exist = os.path.exists(hdf5file)
if hdf5_exist:
os.remove(hdf5file)
@@ -1286,11 +1259,9 @@ class BayesInference:
"""
- MetaModel = self.MetaModel
+ MetaModel = self.engine.MetaModel
Model = self.engine.Model
- out_dir = f'Outputs_Bayes_{Model.name}_{self.name}'
opt_sigma = self.Discrepancy.opt_sigma
-
# -------- Find MAP and run MetaModel and origModel --------
# Compute the MAP
if self.max_a_posteriori.lower() == 'mean':
@@ -1301,6 +1272,8 @@ class BayesInference:
map_theta = Posterior_df.mean(axis=0).reshape(
(1, MetaModel.n_params))
else:
+ # TODO: here just a fix, no previous mention of Posterior_df!
+ Posterior_df = None
map_theta = stats.mode(Posterior_df.values, axis=0)[0]
# Prin report
print("\nPoint estimator:\n", map_theta[0])
@@ -1324,10 +1297,9 @@ class BayesInference:
Marker = 'x'
# Create a PdfPages object
- pdf = PdfPages(f'./{out_dir}MAP_PCE_vs_Model_{self.name}.pdf')
+ pdf = PdfPages(f'./{self.out_dir}MAP_PCE_vs_Model_{self.name}.pdf')
fig = plt.figure()
for i, key in enumerate(Model.Output.names):
-
y_val = map_orig_model[key]
y_pce_val = map_metamodel_mean[key]
y_pce_val_std = map_metamodel_std[key]
@@ -1338,13 +1310,13 @@ class BayesInference:
plt.plot(
x_values, y_pce_val[i], color=Color[i], lw=2.0,
marker=Marker, linestyle='--', label='$Y_{MAP}^{PCE}$'
- )
+ )
# plot the confidence interval
plt.fill_between(
- x_values, y_pce_val[i] - 1.96*y_pce_val_std[i],
- y_pce_val[i] + 1.96*y_pce_val_std[i],
+ x_values, y_pce_val[i] - 1.96 * y_pce_val_std[i],
+ y_pce_val[i] + 1.96 * y_pce_val_std[i],
color=Color[i], alpha=0.15
- )
+ )
# Calculate the adjusted R_squared and RMSE
R2 = r2_score(y_pce_val.reshape(-1, 1), y_val.reshape(-1, 1))
@@ -1359,7 +1331,7 @@ class BayesInference:
fig.canvas.draw()
p = leg.get_window_extent().inverse_transformed(ax.transAxes)
ax.text(
- p.p0[1]-0.05, p.p1[1]-0.25,
+ p.p0[1] - 0.05, p.p1[1] - 0.25,
f'RMSE = {rmse:.3f}\n$R^2$ = {R2:.3f}',
transform=ax.transAxes, color='black',
bbox=dict(facecolor='none', edgecolor='black',
@@ -1375,6 +1347,110 @@ class BayesInference:
pdf.close()
+ def plot_post_params(self, opt_sigma):
+ """
+ Plots the multivar. posterior parameter distribution.
+
+
+ Parameters
+ ----------
+ opt_sigma : string
+ Type of uncertainty description available.
+
+ Returns
+ -------
+ None.
+
+ """
+ par_names = self.engine.ExpDesign.par_names
+ if opt_sigma != "B":
+ par_names.extend(
+ [self.Discrepancy.InputDisc.Marginals[i].name for i
+ in range(len(self.Discrepancy.InputDisc.Marginals))]
+ )
+ # Pot with corner
+ figPosterior = corner.corner(self.posterior_df.to_numpy(),
+ labels=par_names,
+ quantiles=[0.15, 0.5, 0.85],
+ show_titles=True,
+ title_fmt=self.corner_title_fmt,
+ labelpad=0.2,
+ use_math_text=True,
+ title_kwargs={"fontsize": 28},
+ plot_datapoints=False,
+ plot_density=False,
+ fill_contours=True,
+ smooth=0.5,
+ smooth1d=0.5)
+
+ # Loop over axes and set x limits
+ if opt_sigma == "B":
+ axes = np.array(figPosterior.axes).reshape(
+ (len(par_names), len(par_names))
+ )
+ for yi in range(len(par_names)):
+ ax = axes[yi, yi]
+ ax.set_xlim(self.engine.ExpDesign.bound_tuples[yi])
+ for xi in range(yi):
+ ax = axes[yi, xi]
+ ax.set_xlim(self.engine.ExpDesign.bound_tuples[xi])
+ plt.close()
+
+ # Turn off gridlines
+ for ax in figPosterior.axes:
+ ax.grid(False)
+
+ if self.emulator:
+ plotname = f'/Posterior_Dist_{self.engine.Model.name}_emulator'
+ else:
+ plotname = f'/Posterior_Dist_{self.engine.Model.name}'
+
+ figPosterior.set_size_inches((24, 16))
+ figPosterior.savefig(f'./{self.out_dir}{plotname}.pdf',
+ bbox_inches='tight')
+
+ plt.clf()
+
+ def plot_log_BME(self):
+ """
+ Plots the log_BME if bootstrap is active.
+
+ Returns
+ -------
+ None.
+
+ """
+
+ # Computing the TOM performance
+ self.log_BME_tom = stats.chi2.rvs(
+ self.n_tot_measurement, size=self.log_BME.shape[0]
+ )
+
+ fig, ax = plt.subplots()
+ sns.kdeplot(self.log_BME_tom, ax=ax, color="green", shade=True)
+ sns.kdeplot(
+ self.log_BME, ax=ax, color="blue", shade=True,
+ label='Model BME')
+
+ ax.set_xlabel('log$_{10}$(BME)')
+ ax.set_ylabel('Probability density')
+
+ legend_elements = [
+ Patch(facecolor='green', edgecolor='green', label='TOM BME'),
+ Patch(facecolor='blue', edgecolor='blue', label='Model BME')
+ ]
+ ax.legend(handles=legend_elements)
+
+ if self.emulator:
+ plotname = f'/BME_hist_{self.engine.Model.name}_emulator'
+ else:
+ plotname = f'/BME_hist_{self.engine.Model.name}'
+
+ plt.savefig(f'./{self.out_dir}{plotname}.pdf', bbox_inches='tight')
+
+ plt.show()
+ plt.clf()
+
# -------------------------------------------------------------------------
def _plot_post_predictive(self):
"""
@@ -1387,7 +1463,6 @@ class BayesInference:
"""
Model = self.engine.Model
- out_dir = f'Outputs_Bayes_{Model.name}_{self.name}'
# Plot the posterior predictive
for out_idx, out_name in enumerate(Model.Output.names):
fig, ax = plt.subplots()
@@ -1396,11 +1471,10 @@ class BayesInference:
# --- Read prior and posterior predictive ---
if self.inference_method == 'rejection' and \
- self.name.lower() != 'valid':
+ self.name.lower() == 'calib':
# --- Prior ---
# Load posterior predictive
- f = h5py.File(
- f'{out_dir}/priorPredictive.hdf5', 'r+')
+ f = h5py.File(f'{self.out_dir}/priorPredictive.hdf5', 'r+')
try:
x_coords = np.array(f[f"x_values/{out_name}"])
@@ -1409,10 +1483,8 @@ class BayesInference:
X_values = np.repeat(x_coords, 10000)
- prior_pred_df = {}
- prior_pred_df[x_key] = X_values
- prior_pred_df[out_name] = np.array(
- f[f"EDY/{out_name}"])[:10000].flatten('F')
+ prior_pred_df = {x_key: X_values, out_name: np.array(
+ f[f"EDY/{out_name}"])[:10000].flatten('F')}
prior_pred_df = pd.DataFrame(prior_pred_df)
tags_post = ['prior'] * len(prior_pred_df)
@@ -1422,16 +1494,13 @@ class BayesInference:
f.close()
# --- Posterior ---
- f = h5py.File(f"{out_dir}/postPredictive.hdf5", 'r+')
+ f = h5py.File(f"{self.out_dir}/postPredictive.hdf5", 'r+')
X_values = np.repeat(
x_coords, np.array(f[f"EDY/{out_name}"]).shape[0])
- post_pred_df = {}
- post_pred_df[x_key] = X_values
- post_pred_df[out_name] = np.array(
- f[f"EDY/{out_name}"]).flatten('F')
-
+ post_pred_df = {x_key: X_values, out_name: np.array(
+ f[f"EDY/{out_name}"]).flatten('F')}
post_pred_df = pd.DataFrame(post_pred_df)
tags_post = ['posterior'] * len(post_pred_df)
@@ -1459,7 +1528,7 @@ class BayesInference:
ax.errorbar(
x_coords, obs_data[out_name].values,
- yerr=1.96*self.measurement_error[out_name],
+ yerr=1.96 * self.measurement_error[out_name],
ecolor='g', fmt=' ', zorder=-1)
# Add labels to the legend
@@ -1477,7 +1546,7 @@ class BayesInference:
else:
# Load posterior predictive
- f = h5py.File(f"{out_dir}/postPredictive.hdf5", 'r+')
+ f = h5py.File(f"{self.out_dir}/postPredictive.hdf5", 'r+')
try:
x_coords = np.array(f[f"x_values/{out_name}"])
@@ -1492,7 +1561,7 @@ class BayesInference:
x_coords, mu, marker='o', color='b',
label='Mean Post. Predictive')
plt.fill_between(
- x_coords, mu-1.96*std, mu+1.96*std, color='b',
+ x_coords, mu - 1.96 * std, mu + 1.96 * std, color='b',
alpha=0.15)
# --- Plot Data ---
@@ -1505,7 +1574,7 @@ class BayesInference:
for output in orig_ED_Y:
plt.plot(
x_coords, output, color='grey', alpha=0.15
- )
+ )
# Add labels for axes
plt.xlabel('Time [s]')
@@ -1528,5 +1597,7 @@ class BayesInference:
else:
plotname = f'/Post_Prior_Perd_{Model.name}'
- fig.savefig(f'./{out_dir}{plotname}_{out_name}.pdf',
+ fig.savefig(f'./{self.out_dir}{plotname}_{out_name}.pdf',
bbox_inches='tight')
+
+ plt.clf()
diff --git a/src/bayesvalidrox/bayes_inference/bayes_model_comparison.py b/src/bayesvalidrox/bayes_inference/bayes_model_comparison.py
index 828613556e90ec0c529b91f2592eec148c98136b..a26eaa886d6bfda11c23e0e2a490f9f93fee1c45 100644
--- a/src/bayesvalidrox/bayes_inference/bayes_model_comparison.py
+++ b/src/bayesvalidrox/bayes_inference/bayes_model_comparison.py
@@ -1,6 +1,7 @@
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
+import emcee
import numpy as np
import os
from scipy import stats
@@ -8,6 +9,7 @@ import seaborn as sns
import matplotlib.patches as patches
import matplotlib.colors as mcolors
import matplotlib.pylab as plt
+import pandas as pd
from .bayes_inference import BayesInference
# Load the mplstyle
@@ -27,86 +29,89 @@ class BayesModelComparison:
`True`.
perturbed_data : array of shape (n_bootstrap_itrs, n_obs), optional
User defined perturbed data. The default is `None`.
- n_bootstarp : int
+ n_bootstrap : int
Number of bootstrap iteration. The default is `1000`.
data_noise_level : float
A noise level to perturb the data set. The default is `0.01`.
- just_n_meas : int
- Number of measurements considered for visualization of the
- justifiability results.
"""
def __init__(self, justifiability=True, perturbed_data=None,
- n_bootstarp=1000, data_noise_level=0.01, just_n_meas=2):
+ n_bootstrap=1000, data_noise_level=0.01,
+ use_Bayes_settings = True, emulator = True, out_dir = 'Outputs_Comparison/'):
+ # TODO: check valid ranges of the parameters
+
self.justifiability = justifiability
self.perturbed_data = perturbed_data
- self.n_bootstarp = n_bootstarp
+ self.n_bootstrap = n_bootstrap
self.data_noise_level = data_noise_level
- self.just_n_meas = just_n_meas
-
+ self.use_Bayes_settings = use_Bayes_settings
+ self.emulator = emulator
+ self.out_dir = out_dir
+
+ # Other parameters
+ self.n_meas = None
+ self.BF_data = None
+ self.just_data = None
+ self.BME_dict = None
+ self.set_up = False
+ self.dtype = None
+ self.bayes_dict = None
+ self.just_bayes_dict = None
+ self.model_weights = None
+ self.model_weights_dict = None
+ self.just_model_weights_dict = None
+
+
# --------------------------------------------------------------------------
- def create_model_comparison(self, model_dict, opts_dict):
+ def setup(self, model_dict):
"""
- Starts the two-stage model comparison.
- Stage I: Compare models using Bayes factors.
- Stage II: Compare models via justifiability analysis.
-
- Parameters
- ----------
- model_dict : dict
- A dictionary including the metamodels.
- opts_dict : dict
- A dictionary given the `BayesInference` options.
-
- Example:
-
- >>> opts_bootstrap = {
- "bootstrap": True,
- "n_samples": 10000,
- "Discrepancy": DiscrepancyOpts,
- "emulator": True,
- "plot_post_pred": True
- }
+ Initialize parameters that are needed for all types of model comparison
Returns
-------
- output : dict
- A dictionary containing the objects and the model weights for the
- comparison using Bayes factors and justifiability analysis.
+ None.
"""
+
+ if not isinstance(model_dict, dict):
+ raise Exception("To run model comparsion, you need to pass a "
+ "dictionary of models.")
- # Bayes factor
- bayes_dict_bf, model_weights_dict_bf = self.compare_models(
- model_dict, opts_dict
- )
-
- output = {
- 'Bayes objects BF': bayes_dict_bf,
- 'Model weights BF': model_weights_dict_bf
- }
+ # Extract model names
+ self.model_names = [*model_dict]
- # Justifiability analysis
- if self.justifiability:
- bayes_dict_ja, model_weights_dict_ja = self.compare_models(
- model_dict, opts_dict, justifiability=True
- )
+ # Compute total number of the measurement points
+ # TODO: there could be a different option for this here
+ Engine = list(model_dict.items())[0][1]
+ Engine.Model.read_observation()
+ self.n_meas = Engine.Model.n_obs
- output['Bayes objects JA'] = bayes_dict_ja
- output['Model weights JA'] = model_weights_dict_ja
+ # Find n_bootstrap
+ if self.perturbed_data is not None:
+ self.n_bootstrap = self.perturbed_data.shape[0]
+
+ # Output directory
+ os.makedirs(self.out_dir, exist_ok=True)
+
+ # System settings
+ if os.name == 'nt':
+ print('')
+ print('WARNING: Performing the inference on windows can lead to reduced accuracy!')
+ print('')
+ self.dtype=np.longdouble
+ else:
+ self.dtype=np.float128
- return output
# --------------------------------------------------------------------------
- def compare_models(self, model_dict, opts_dict, justifiability=False):
+ def model_comparison_all(self, model_dict, opts_dict):
"""
- Passes the options to instantiates the BayesInference class for each
- model and passes the options from `opts_dict`. Then, it starts the
- computations.
- It also creates a folder and saves the diagrams, e.g., Bayes factor
- plot, confusion matrix, etc.
+ Perform all three types of model comparison:
+ * Bayes Factors
+ * Model weights
+ * Justifiability analysis
Parameters
----------
@@ -114,50 +119,57 @@ class BayesModelComparison:
A dictionary including the metamodels.
opts_dict : dict
A dictionary given the `BayesInference` options.
- justifiability : bool, optional
- Whether to perform the justifiability analysis. The default is
- `False`.
Returns
-------
- bayes_dict : dict
- A dictionary with `BayesInference` objects.
- model_weights_dict : dict
- A dictionary containing the model weights.
+ results : dict
+ A dictionary that contains the calculated BME values, model weights
+ and confusion matrix
"""
+ self.calc_bayes_factors(model_dict, opts_dict)
+ self.calc_model_weights(model_dict, opts_dict)
+ self.calc_justifiability_analysis(model_dict, opts_dict)
+
+ results = {'BME': self.BME_dict, 'Model weights': self.model_weights_dict,
+ 'Confusion matrix': self.confusion_matrix}
+ return results
+
- if not isinstance(model_dict, dict):
- raise Exception("To run model comparsion, you need to pass a "
- "dictionary of models.")
+ # --------------------------------------------------------------------------
+ def calc_bayes_factors(self, model_dict, opts_dict):
+ """
+ Calculate the BayesFactors for each pair of models in the model_dict
+ with respect to given data.
- # Extract model names
- self.model_names = [*model_dict]
+ Parameters
+ ----------
+ model_dict : dict
+ A dictionary including the metamodels.
+ opts_dict : dict
+ A dictionary given the `BayesInference` options.
- # Compute total number of the measurement points
- Engine = list(model_dict.items())[0][1]
- Engine.Model.read_observation()
- self.n_meas = Engine.Model.n_obs
+ Returns
+ -------
+ None.
+ """
+ # Do the setup
+ if self.n_meas is None:
+ self.setup(model_dict)
+
# ----- Generate data -----
- # Find n_bootstrap
- if self.perturbed_data is None:
- n_bootstarp = self.n_bootstarp
- else:
- n_bootstarp = self.perturbed_data.shape[0]
-
# Create dataset
- justData = self.generate_dataset(
- model_dict, justifiability, n_bootstarp=n_bootstarp)
+ self.BF_data = self.generate_dataset(
+ model_dict, False, n_bootstrap=self.n_bootstrap)
# Run create Interface for each model
- bayes_dict = {}
+ self.bayes_dict = {}
for model in model_dict.keys():
print("-"*20)
print("Bayesian inference of {}.\n".format(model))
-
BayesOpts = BayesInference(model_dict[model])
-
+
# Set BayesInference options
for key, value in opts_dict.items():
if key in BayesOpts.__dict__.keys():
@@ -167,49 +179,147 @@ class BayesModelComparison:
setattr(BayesOpts, key, value)
# Pass justifiability data as perturbed data
- BayesOpts.perturbed_data = justData
- BayesOpts.just_analysis = justifiability
+ BayesOpts.bmc = True
+ BayesOpts.emulator= self.emulator
+ BayesOpts.just_analysis = False
+ BayesOpts.perturbed_data = self.BF_data
- bayes_dict[model] = BayesOpts.create_inference()
+ self.bayes_dict[model] = BayesOpts.create_inference()
print("-"*20)
- # Compute model weights
- BME_Dict = dict()
- for modelName, bayesObj in bayes_dict.items():
- BME_Dict[modelName] = np.exp(bayesObj.log_BME, dtype=np.longdouble)#float128)
+ # Accumulate the BMEs
+ self.BME_dict = dict()
+ for modelName, bayesObj in self.bayes_dict.items():
+ self.BME_dict[modelName] = np.exp(bayesObj.log_BME, dtype=self.dtype)
+
+ # TODO: move the calculation of the Bayes Factors out of the plots to here!
+ # Create kde plot for bayes factors
+ self.plot_bayes_factor(self.BME_dict, 'kde_plot')
+
+
+ def calc_model_weights(self, model_dict, opts_dict):
+ """
+ Calculate the model weights from BME evaluations for Bayes factors.
- # BME correction in BayesInference class
- model_weights = self.cal_model_weight(
- BME_Dict, justifiability, n_bootstarp=n_bootstarp)
+ Parameters
+ ----------
+ model_dict : TYPE
+ DESCRIPTION.
+ opts_dict : TYPE
+ DESCRIPTION.
- # Plot model weights
- if justifiability:
- model_names = self.model_names
- model_names.insert(0, 'Observation')
+ Returns
+ -------
+ None.
+
+ """
+ # Get BMEs via Bayes Factors if not already done so
+ if self.BME_dict is None:
+ self.calc_bayes_factors(model_dict, opts_dict)
+
+ # Calculate the model weights
+ self.model_weights = self.cal_model_weight(
+ self.BME_dict, False, n_bootstrap=self.n_bootstrap)
- # Split the model weights and save in a dict
- list_ModelWeights = np.split(
- model_weights, model_weights.shape[1]/self.n_meas, axis=1)
- model_weights_dict = {key: weights for key, weights in
- zip(model_names, list_ModelWeights)}
+ # Create box plot for model weights
+ self.plot_model_weights(self.model_weights, 'model_weights')
- #self.plot_just_analysis(model_weights_dict)
- else:
- # Create box plot for model weights
- self.plot_model_weights(model_weights, 'model_weights')
- # Create kde plot for bayes factors
- self.plot_bayes_factor(BME_Dict, 'kde_plot')
+ # -------------------------------------------------------------------------
+ def calc_justifiability_analysis(self, model_dict, opts_dict):
+ """
+ Perform justifiability analysis by calculating the confusion matrix
+
+ Parameters
+ ----------
+ model_dict : dict
+ A dictionary including the metamodels.
+ opts_dict : dict
+ A dictionary given the `BayesInference` options.
+
+ Returns
+ -------
+ confusion_matrix: dict
+ The averaged confusion matrix
+
+ """
+ # Do setup
+ if self.n_meas is None:
+ self.setup(model_dict)
+
+ # Extend model names
+ model_names = self.model_names
+ if model_names[0]!= 'Observation':
+ model_names.insert(0, 'Observation')
+
+ # Generate data
+ # TODO: generate the datset only if it does not exist yet
+ self.just_data = self.generate_dataset(
+ model_dict, True, n_bootstrap=self.n_bootstrap)
+
+ # Run inference for each model if this is not available
+ if self.just_bayes_dict is None:
+ self.just_bayes_dict = {}
+ for model in model_dict.keys():
+ print("-"*20)
+ print("Bayesian inference of {}.\n".format(model))
+ BayesOpts = BayesInference(model_dict[model])
+
+ # Set BayesInference options
+ for key, value in opts_dict.items():
+ if key in BayesOpts.__dict__.keys():
+ if key == "Discrepancy" and isinstance(value, dict):
+ setattr(BayesOpts, key, value[model])
+ else:
+ setattr(BayesOpts, key, value)
+
+ # Pass justifiability data as perturbed data
+ BayesOpts.bmc = True
+ BayesOpts.emulator= self.emulator
+ BayesOpts.just_analysis = True
+ BayesOpts.perturbed_data = self.just_data
+
+ self.just_bayes_dict[model] = BayesOpts.create_inference()
+ print("-"*20)
+
+ # Compute model weights
+ self.BME_dict = dict()
+ for modelName, bayesObj in self.bayes_dict.items():
+ self.BME_dict[modelName] = np.exp(bayesObj.log_BME, dtype=self.dtype)
- # Store model weights in a dict
- model_weights_dict = {key: weights for key, weights in
- zip(self.model_names, model_weights)}
+ # BME correction in BayesInference class
+ just_model_weights = self.cal_model_weight(
+ self.BME_dict, True, n_bootstrap=self.n_bootstrap)
+
+ # Split the model weights and save in a dict
+ list_ModelWeights = np.split(
+ just_model_weights, self.model_weights.shape[1]/self.n_meas, axis=1)
+ self.just_model_weights_dict = {key: weights for key, weights in
+ zip(model_names, list_ModelWeights)}
+
+ # Confusion matrix over all measurement points
+ cf_m = pd.DataFrame()
+ cf_m['Generated by'] = model_names
+ for i in range(len(model_names)):
+ # Ignore 'Observation', this not in the model_weights_dict
+ # TODO: how to change the code so that it is included as well?
+ if i==0:
+ continue
+ avg = []
+ for n in model_names:
+ avg.append(np.sum(self.just_model_weights_dict[n][i-1]))
+
+ # Norm to sum to 1 for each 'Generated by' row
+ cf_m[model_names[i]] = avg/self.n_meas
+ self.confusion_matrix = cf_m
+
+ # Plot model weights
+ self.plot_just_analysis()
- return bayes_dict, model_weights_dict
# -------------------------------------------------------------------------
def generate_dataset(self, model_dict, justifiability=False,
- n_bootstarp=1):
+ n_bootstrap=1):
"""
Generates the perturbed data set for the Bayes factor calculations and
the data set for the justifiability analysis.
@@ -221,7 +331,7 @@ class BayesModelComparison:
bool, optional
Whether to perform the justifiability analysis. The default is
`False`.
- n_bootstarp : int, optional
+ n_bootstrap : int, optional
Number of bootstrap iterations. The default is `1`.
Returns
@@ -238,27 +348,28 @@ class BayesModelComparison:
# Perturb observations for Bayes Factor
if self.perturbed_data is None:
self.perturbed_data = self.__perturb_data(
- Engine.Model.observations, out_names, n_bootstarp,
+ Engine.Model.observations, out_names, n_bootstrap,
noise_level=self.data_noise_level)
# Only for Bayes Factor
if not justifiability:
- return self.perturbed_data
+ return self.perturbed_data # TODO: why return this as self... and the other one not? Is this used again?
# Evaluate metamodel
runs = {}
- for key, metaModel in model_dict.items():
- y_hat, _ = metaModel.eval_metamodel(nsamples=n_bootstarp)
+ for key, metaModel in model_dict.items(): # TODO: add check for emulator vs model
+ y_hat, _ = metaModel.eval_metamodel(nsamples=n_bootstrap)
runs[key] = y_hat
# Generate data
- for i in range(n_bootstarp):
- y_data = self.perturbed_data[i].reshape(1, -1)
- justData = np.tril(np.repeat(y_data, y_data.shape[1], axis=0))
+ for i in range(n_bootstrap):
+ y_data = self.perturbed_data[i].reshape(1, -1)# makes every entry in self.perturbed_data 2D by adding one dim outside
+ justData = np.tril(np.repeat(y_data, y_data.shape[1], axis=0)) # Lower triangle matrix from repeats of y_data
+ # TODO: why triangle matrix here?
# Use surrogate runs for data-generating process
for key, metaModel in model_dict.items():
model_data = np.array(
- [runs[key][out][i] for out in out_names]).reshape(y_data.shape)
+ [runs[key][out][i] for out in out_names]).reshape(y_data.shape) # reshapes model runs to match y_data
justData = np.vstack((
justData,
np.tril(np.repeat(model_data, model_data.shape[1], axis=0))
@@ -276,7 +387,7 @@ class BayesModelComparison:
# -------------------------------------------------------------------------
def __perturb_data(self, data, output_names, n_bootstrap, noise_level):
"""
- Returns an array with n_bootstrap_itrs rowsof perturbed data.
+ Returns an array with n_bootstrap_itrs rows of perturbed data.
The first row includes the original observation data.
If `self.bayes_loocv` is True, a 2d-array will be returned with
repeated rows and zero diagonal entries.
@@ -318,13 +429,13 @@ class BayesModelComparison:
return final_data
# -------------------------------------------------------------------------
- def cal_model_weight(self, BME_Dict, justifiability=False, n_bootstarp=1):
+ def cal_model_weight(self, BME_dict, justifiability=False, n_bootstrap=1):
"""
Normalize the BME (Asumption: Model Prior weights are equal for models)
Parameters
----------
- BME_Dict : dict
+ BME_dict : dict
A dictionary containing the BME values.
Returns
@@ -334,12 +445,12 @@ class BayesModelComparison:
"""
# Stack the BME values for all models
- all_BME = np.vstack(list(BME_Dict.values()))
+ all_BME = np.vstack(list(BME_dict.values()))
if justifiability:
# Compute expected log_BME for justifiabiliy analysis
all_BME = all_BME.reshape(
- all_BME.shape[0], -1, n_bootstarp).mean(axis=2)
+ all_BME.shape[0], -1, n_bootstrap).mean(axis=2)
# Model weights
model_weights = np.divide(all_BME, np.nansum(all_BME, axis=0))
@@ -362,16 +473,16 @@ class BayesModelComparison:
None.
"""
-
- directory = 'Outputs_Comparison/'
- os.makedirs(directory, exist_ok=True)
+ model_weights_dict = self.just_model_weights_dict
Color = [*mcolors.TABLEAU_COLORS]
names = [*model_weights_dict]
+ # Plot weights for each 'Generated by'
model_names = [model.replace('_', '$-$') for model in self.model_names]
for name in names:
fig, ax = plt.subplots()
for i, model in enumerate(model_names[1:]):
+ #print(model, i)
plt.plot(list(range(1, self.n_meas+1)),
model_weights_dict[name][i],
color=Color[i], marker='o',
@@ -384,13 +495,12 @@ class BayesModelComparison:
ax.set_xticks(list(range(1, self.n_meas+1)))
plt.legend(loc="best")
fig.savefig(
- f'{directory}modelWeights_{name}.svg', bbox_inches='tight'
+ f'{self.out_dir}modelWeights_{name}.svg', bbox_inches='tight'
)
plt.close()
- # Confusion matrix for some measurement points
- epsilon = 1 if self.just_n_meas != 1 else 0
- for index in range(0, self.n_meas+epsilon, self.just_n_meas):
+ # Confusion matrix for each measurement point
+ for index in range(0, self.n_meas):
weights = np.array(
[model_weights_dict[key][:, index] for key in model_weights_dict]
)
@@ -405,10 +515,26 @@ class BayesModelComparison:
g.set_xlabel(r"\textbf{Data generated by:}", labelpad=15)
g.set_ylabel(r"\textbf{Model weight for:}", labelpad=15)
g.figure.savefig(
- f"{directory}confusionMatrix_ND_{index+1}.pdf",
+ f"{self.out_dir}confusionMatrix_ND_{index+1}.pdf",
bbox_inches='tight'
)
plt.close()
+
+ # Plot the averaged confusion matrix
+ out_names = names[1:]
+ cf = self.confusion_matrix[out_names].to_numpy()
+ g = sns.heatmap(cf.T, annot=True, cmap='Blues', xticklabels=model_names,
+ yticklabels=model_names[1:], annot_kws={"size": 24})
+ g.xaxis.tick_top()
+ g.xaxis.set_label_position('top')
+ g.set_xlabel(r"\textbf{Data generated by:}", labelpad=15)
+ g.set_ylabel(r"\textbf{Model weight for:}", labelpad=15)
+ g.figure.savefig(
+ f"{self.out_dir}confusionMatrix_full.pdf",
+ bbox_inches='tight'
+ )
+ plt.close()
+
# -------------------------------------------------------------------------
def plot_model_weights(self, model_weights, plot_name):
@@ -428,13 +554,9 @@ class BayesModelComparison:
None.
"""
- font_size = 40
- # mkdir for plots
- directory = 'Outputs_Comparison/'
- os.makedirs(directory, exist_ok=True)
-
# Create figure
fig, ax = plt.subplots()
+ font_size = 40
# Filter data using np.isnan
mask = ~np.isnan(model_weights.T)
@@ -462,44 +584,35 @@ class BayesModelComparison:
for median in bp['medians']:
median.set(color='#b2df8a', linewidth=2)
- # change the style of fliers and their fill
- # for flier in bp['fliers']:
- # flier.set(marker='o', color='#e7298a', alpha=0.75)
-
- # Custom x-axis labels
+ # Customize the axes
model_names = [model.replace('_', '$-$') for model in self.model_names]
ax.set_xticklabels(model_names)
-
ax.set_ylabel('Weight', fontsize=font_size)
-
- # Title
- plt.title('Posterior Model Weights')
-
- # Set y lim
ax.set_ylim((-0.05, 1.05))
-
- # Set size of the ticks
for t in ax.get_xticklabels():
t.set_fontsize(font_size)
for t in ax.get_yticklabels():
t.set_fontsize(font_size)
+ # Title
+ plt.title('Posterior Model Weights')
+
# Save the figure
fig.savefig(
- f'./{directory}{plot_name}.pdf', bbox_inches='tight'
+ f'./{self.out_dir}{plot_name}.pdf', bbox_inches='tight'
)
plt.close()
# -------------------------------------------------------------------------
- def plot_bayes_factor(self, BME_Dict, plot_name=''):
+ def plot_bayes_factor(self, BME_dict, plot_name=''):
"""
Plots the Bayes factor distibutions in a :math:`N_m \\times N_m`
matrix, where :math:`N_m` is the number of the models.
Parameters
----------
- BME_Dict : dict
+ BME_dict : dict
A dictionary containing the BME values of the models.
plot_name : str, optional
Plot name. The default is ''.
@@ -509,16 +622,11 @@ class BayesModelComparison:
None.
"""
-
+ # Plot setup
font_size = 40
-
- # mkdir for plots
- directory = 'Outputs_Comparison/'
- os.makedirs(directory, exist_ok=True)
-
Colors = ["blue", "green", "gray", "brown"]
- model_names = list(BME_Dict.keys())
+ model_names = list(BME_dict.keys())
nModels = len(model_names)
# Plots
@@ -540,7 +648,7 @@ class BayesModelComparison:
# Null hypothesis: key_j is the better model
BayesFactor = np.log10(
- np.divide(BME_Dict[key_i], BME_Dict[key_j])
+ np.divide(BME_dict[key_i], BME_dict[key_j])
)
# sns.kdeplot(BayesFactor, ax=ax, color=Colors[i], shade=True)
@@ -633,10 +741,8 @@ class BayesModelComparison:
fontsize=fsize, color=Colors[i],
transform=ax.transAxes)
- # Defining custom 'ylim' values.
+ # Customize axes
custom_ylim = (0, 1.05)
-
- # Setting the values for all axes.
plt.setp(axes, ylim=custom_ylim)
# set labels
@@ -648,7 +754,7 @@ class BayesModelComparison:
plt.subplots_adjust(wspace=0.2, hspace=0.1)
plt.savefig(
- f'./{directory}Bayes_Factor{plot_name}.pdf', bbox_inches='tight'
+ f'./{self.out_dir}Bayes_Factor{plot_name}.pdf', bbox_inches='tight'
)
plt.close()
diff --git a/src/bayesvalidrox/bayes_inference/discrepancy.py b/src/bayesvalidrox/bayes_inference/discrepancy.py
index fff32a2500ae20b3667c7b0ec2cc85c1da614688..b3c235ebeb6d6ae9e109ca862cc522cc21efb45e 100644
--- a/src/bayesvalidrox/bayes_inference/discrepancy.py
+++ b/src/bayesvalidrox/bayes_inference/discrepancy.py
@@ -36,7 +36,7 @@ class Discrepancy:
* Option B: With unknown redidual covariance matrix \\(\\Sigma\\),
paramethrized as \\(\\Sigma(\\theta_{\\epsilon})=\\sigma^2 \\textbf{I}_
{N_{out}}\\) with unknown residual variances \\(\\sigma^2\\).
- This term will be jointly infer with the uncertain input parameters. For
+ This term will be jointly infered with the uncertain input parameters. For
the inversion, you need to define a prior marginal via `Input` class. Note
that \\(\\sigma^2\\) is only a single scalar multiplier for the diagonal
entries of the covariance matrix \\(\\Sigma\\).
@@ -58,10 +58,17 @@ class Discrepancy:
"""
def __init__(self, InputDisc='', disc_type='Gaussian', parameters=None):
+ # Set the values
self.InputDisc = InputDisc
self.disc_type = disc_type
self.parameters = parameters
-
+
+ # Other inits
+ self.ExpDesign = None
+ self.n_samples = None
+ self.sigma2_prior = None
+ self.name = None
+ self.opt_sigma = None # This will be set in the inference class and used in mcmc
# -------------------------------------------------------------------------
def get_sample(self, n_samples):
"""
@@ -87,6 +94,11 @@ class Discrepancy:
# Create and store BoundTuples
self.ExpDesign = ExpDesigns(self.InputDisc)
self.ExpDesign.sampling_method = 'random'
+
+ # TODO: why does it call 'generate_ED' instead of 'generate_samples?
+ # ExpDesign.bound_tuples, onp_sigma, prior_space needed from the outside
+ # Discrepancy opt_sigma, InputDisc needed from the outside
+ # TODO: opt_sigma not defined here, but called from the outside??
self.ExpDesign.generate_ED(
n_samples, max_pce_deg=1
)
diff --git a/src/bayesvalidrox/bayes_inference/mcmc.py b/src/bayesvalidrox/bayes_inference/mcmc.py
index fe22a152f117aab7023bfe6592ce3a48bb0b3aec..f4d1524d3acbd2e09866042a47670c1f0ffad213 100755
--- a/src/bayesvalidrox/bayes_inference/mcmc.py
+++ b/src/bayesvalidrox/bayes_inference/mcmc.py
@@ -15,6 +15,85 @@ import shutil
os.environ["OMP_NUM_THREADS"] = "1"
+# -------------------------------------------------------------------------
+def _check_ranges(theta, ranges): # TODO: this is a replica of exp_designs.check_ranges
+ """
+ This function checks if theta lies in the given ranges.
+
+ Parameters
+ ----------
+ theta : array
+ Proposed parameter set.
+ ranges : nested list
+ List of the praremeter ranges.
+
+ Returns
+ -------
+ c : bool
+ If it lies in the given range, it return True else False.
+
+ """
+ c = True
+ # traverse in the list1
+ for i, bounds in enumerate(ranges):
+ x = theta[i]
+ # condition check
+ if x < bounds[0] or x > bounds[1]:
+ c = False
+ return c
+ return c
+
+# -------------------------------------------------------------------------
+def gelman_rubin(chain, return_var=False):
+ """
+ The potential scale reduction factor (PSRF) defined by the variance
+ within one chain, W, with the variance between chains B.
+ Both variances are combined in a weighted sum to obtain an estimate of
+ the variance of a parameter \\( \\theta \\).The square root of the
+ ratio of this estimates variance to the within chain variance is called
+ the potential scale reduction.
+ For a well converged chain it should approach 1. Values greater than
+ 1.1 typically indicate that the chains have not yet fully converged.
+
+ Source: http://joergdietrich.github.io/emcee-convergence.html
+
+ https://github.com/jwalton3141/jwalton3141.github.io/blob/master/assets/posts/ESS/rwmh.py
+
+ Parameters
+ ----------
+ chain : array (n_walkers, n_steps, n_params)
+ The emcee ensamples.
+
+ Returns
+ -------
+ R_hat : float
+ The Gelman-Robin values.
+
+ """
+ chain = np.array(chain)
+ m_chains, n_iters = chain.shape[:2]
+
+ # Calculate between-chain variance
+ θb = np.mean(chain, axis=1)
+ θbb = np.mean(θb, axis=0)
+ B_over_n = ((θbb - θb)**2).sum(axis=0)
+ B_over_n /= (m_chains - 1)
+
+ # Calculate within-chain variances
+ ssq = np.var(chain, axis=1, ddof=1)
+ W = np.mean(ssq, axis=0)
+
+ # (over) estimate of variance
+ var_θ = W * (n_iters - 1) / n_iters + B_over_n
+
+ if return_var:
+ return var_θ
+ else:
+ # The square root of the ratio of this estimates variance to the
+ # within chain variance
+ R_hat = np.sqrt(var_θ / W)
+ return R_hat
+
class MCMC:
"""
A class for bayesian inference via a Markov-Chain Monte-Carlo (MCMC)
@@ -41,65 +120,67 @@ class MCMC:
"""
def __init__(self, BayesOpts):
-
+ # Inputs
self.BayesOpts = BayesOpts
+
+ # Param inits
+ self.counter = 0
+ self.observation = None
+ self.total_sigma2 = None
+
+ # Get general params from BayesOpts
+ self.out_dir = self.BayesOpts.out_dir
+
+ # Get MCMC parameters ftom BayesOpts
+ pars = self.BayesOpts.mcmc_params
+ self.initsamples = pars['init_samples']
+ if isinstance(self.initsamples, pd.DataFrame):
+ self.initsamples = self.initsamples.values
+ self.nsteps = int(pars['n_steps'])
+ self.nwalkers = int(pars['n_walkers'])
+ self.nburn = pars['n_burn']
+ self.moves = pars['moves']
+ self.mp = pars['multiprocessing']
+ self.verbose = pars['verbose']
def run_sampler(self, observation, total_sigma2):
+ """
+ Run the MCMC sampler for the given observations and stdevs.
+
+ Parameters
+ ----------
+ observation : TYPE
+ DESCRIPTION.
+ total_sigma2 : TYPE
+ DESCRIPTION.
+ Returns
+ -------
+ Posterior_df : TYPE
+ DESCRIPTION.
+
+ """
+ # Get init values
BayesObj = self.BayesOpts
- MetaModel = BayesObj.engine.MetaModel
- Model = BayesObj.engine.Model
Discrepancy = self.BayesOpts.Discrepancy
- n_cpus = Model.n_cpus
- priorDist = BayesObj.engine.ExpDesign.JDist
- ndim = MetaModel.n_params
- self.counter = 0
- output_dir = f'Outputs_Bayes_{Model.name}_{self.BayesOpts.name}'
- if not os.path.exists(output_dir):
- os.makedirs(output_dir)
+ n_cpus = BayesObj.engine.Model.n_cpus
+ ndim = BayesObj.engine.MetaModel.n_params
+ if not os.path.exists(self.out_dir):
+ os.makedirs(self.out_dir)
+ # Save inputs
self.observation = observation
self.total_sigma2 = total_sigma2
- # Unpack mcmc parameters given to BayesObj.mcmc_params
- self.initsamples = None
- self.nwalkers = 100
- self.nburn = 200
- self.nsteps = 100000
- self.moves = None
- self.mp = False
- self.verbose = False
-
- # Extract initial samples
- if 'init_samples' in BayesObj.mcmc_params:
- self.initsamples = BayesObj.mcmc_params['init_samples']
- if isinstance(self.initsamples, pd.DataFrame):
- self.initsamples = self.initsamples.values
-
- # Extract number of steps per walker
- if 'n_steps' in BayesObj.mcmc_params:
- self.nsteps = int(BayesObj.mcmc_params['n_steps'])
- # Extract number of walkers (chains)
- if 'n_walkers' in BayesObj.mcmc_params:
- self.nwalkers = int(BayesObj.mcmc_params['n_walkers'])
- # Extract moves
- if 'moves' in BayesObj.mcmc_params:
- self.moves = BayesObj.mcmc_params['moves']
- # Extract multiprocessing
- if 'multiprocessing' in BayesObj.mcmc_params:
- self.mp = BayesObj.mcmc_params['multiprocessing']
- # Extract verbose
- if 'verbose' in BayesObj.mcmc_params:
- self.verbose = BayesObj.mcmc_params['verbose']
-
# Set initial samples
np.random.seed(0)
if self.initsamples is None:
try:
- initsamples = priorDist.sample(self.nwalkers).T
+ initsamples = BayesObj.engine.ExpDesign.JDist.sample(self.nwalkers).T
+ initsamples = np.swapaxes(np.array([initsamples]),0,1) # TODO: test if this still works with multiple input dists
except:
# when aPCE selected - gaussian kernel distribution
- inputSamples = MetaModel.ExpDesign.raw_data.T
+ inputSamples = self.BayesOpts.engine.ExpDesign.raw_data.T
random_indices = np.random.choice(
len(inputSamples), size=self.nwalkers, replace=False
)
@@ -125,16 +206,14 @@ class MCMC:
initsamples[:, idx_dim] = dist.rvs(size=self.nwalkers)
# Update lower and upper
- MetaModel.ExpDesign.bound_tuples = bound_tuples
+ BayesObj.engine.MetaModel.ExpDesign.bound_tuples = bound_tuples
# Check if sigma^2 needs to be inferred
- if Discrepancy.opt_sigma != 'B':
+ if Discrepancy.opt_sigma != 'B': # TODO: why !='B'?
sigma2_samples = Discrepancy.get_sample(self.nwalkers)
# Update initsamples
initsamples = np.hstack((initsamples, sigma2_samples))
-
- # Update ndim
ndim = initsamples.shape[1]
# Discrepancy bound
@@ -146,10 +225,8 @@ class MCMC:
print("\n>>>> Bayesian inference with MCMC for "
f"{self.BayesOpts.name} started. <<<<<<")
- # Set up the backend
- filename = f"{output_dir}/emcee_sampler.h5"
- backend = emcee.backends.HDFBackend(filename)
- # Clear the backend in case the file already exists
+ # Set up the backend and clear it in case the file already exists
+ backend = emcee.backends.HDFBackend(f"{self.out_dir}/emcee_sampler.h5")
backend.reset(self.nwalkers, ndim)
# Define emcee sampler
@@ -176,8 +253,8 @@ class MCMC:
)
# Reset sampler
- sampler.reset()
pos = pos.coords
+ sampler.reset()
else:
pos = initsamples
@@ -252,13 +329,13 @@ class MCMC:
# output current autocorrelation estimate
if self.verbose:
print(f"Mean autocorr. time estimate: {np.nanmean(tau):.3f}")
- list_gr = np.round(self.gelman_rubin(sampler.chain), 3)
+ list_gr = np.round(gelman_rubin(sampler.chain), 3)
print("Gelman-Rubin Test*: ", list_gr)
# check convergence
converged = np.all(tau*autocorreverynsteps < sampler.iteration)
converged &= np.all(np.abs(tauold - tau) / tau < 0.01)
- converged &= np.all(self.gelman_rubin(sampler.chain) < 1.1)
+ converged &= np.all(gelman_rubin(sampler.chain) < 1.1)
if converged:
break
@@ -277,7 +354,7 @@ class MCMC:
thin = int(0.5*np.nanmin(tau)) if int(0.5*np.nanmin(tau)) != 0 else 1
finalsamples = sampler.get_chain(discard=burnin, flat=True, thin=thin)
acc_fr = np.nanmean(sampler.acceptance_fraction)
- list_gr = np.round(self.gelman_rubin(sampler.chain[:, burnin:]), 3)
+ list_gr = np.round(gelman_rubin(sampler.chain[:, burnin:]), 3)
# Print summary
print('\n')
@@ -307,7 +384,7 @@ class MCMC:
# Plot traces
if self.verbose and self.nsteps < 10000:
- pdf = PdfPages(output_dir+'/traceplots.pdf')
+ pdf = PdfPages(self.out_dir+'/traceplots.pdf')
fig = plt.figure()
for parIdx in range(ndim):
# Set up the axes with gridspec
@@ -334,7 +411,6 @@ class MCMC:
# Destroy the current plot
plt.clf()
-
pdf.close()
# plot development of autocorrelation estimate
@@ -348,33 +424,9 @@ class MCMC:
plt.ylim(0, np.nanmax(taus)+0.1*(np.nanmax(taus)-np.nanmin(taus)))
plt.xlabel("number of steps")
plt.ylabel(r"mean $\hat{\tau}$")
- fig1.savefig(f"{output_dir}/autocorrelation_time.pdf",
+ fig1.savefig(f"{self.out_dir}/autocorrelation_time.pdf",
bbox_inches='tight')
- # logml_dict = self.marginal_llk_emcee(sampler, self.nburn, logp=None,
- # maxiter=5000)
- # print('\nThe Bridge Sampling Estimation is "
- # f"{logml_dict['logml']:.5f}.')
-
- # # Posterior-based expectation of posterior probablity
- # postExpPostLikelihoods = np.mean(sampler.get_log_prob(flat=True)
- # [self.nburn*self.nwalkers:])
-
- # # Posterior-based expectation of prior densities
- # postExpPrior = np.mean(self.log_prior(emcee_trace.T))
-
- # # Posterior-based expectation of likelihoods
- # postExpLikelihoods_emcee = postExpPostLikelihoods - postExpPrior
-
- # # Calculate Kullback-Leibler Divergence
- # KLD_emcee = postExpLikelihoods_emcee - logml_dict['logml']
- # print("Kullback-Leibler divergence: %.5f"%KLD_emcee)
-
- # # Information Entropy based on Entropy paper Eq. 38
- # infEntropy_emcee = logml_dict['logml'] - postExpPrior -
- # postExpLikelihoods_emcee
- # print("Information Entropy: %.5f" %infEntropy_emcee)
-
Posterior_df = pd.DataFrame(finalsamples, columns=par_names)
return Posterior_df
@@ -397,8 +449,7 @@ class MCMC:
returned otherwise an array.
"""
-
- MetaModel = self.BayesOpts.MetaModel
+ MetaModel = self.BayesOpts.engine.MetaModel
Discrepancy = self.BayesOpts.Discrepancy
# Find the number of sigma2 parameters
@@ -417,7 +468,7 @@ class MCMC:
for i in range(nsamples):
# Check if the sample is within the parameters' range
- if self._check_ranges(theta[i], params_range):
+ if _check_ranges(theta[i], params_range):
# Check if all dists are uniform, if yes priors are equal.
if all(MetaModel.input_obj.Marginals[i].dist_type == 'uniform'
for i in range(MetaModel.n_params)):
@@ -429,7 +480,7 @@ class MCMC:
# Check if bias term needs to be inferred
if Discrepancy.opt_sigma != 'B':
- if self._check_ranges(theta[i, -n_sigma2:],
+ if _check_ranges(theta[i, -n_sigma2:],
disc_bound_tuples):
if all('unif' in disc_marginals[i].dist_type for i in
range(Discrepancy.ExpDesign.ndim)):
@@ -463,21 +514,20 @@ class MCMC:
"""
BayesOpts = self.BayesOpts
- MetaModel = BayesOpts.MetaModel
+ MetaModel = BayesOpts.engine.MetaModel
Discrepancy = self.BayesOpts.Discrepancy
# Find the number of sigma2 parameters
if Discrepancy.opt_sigma != 'B':
disc_bound_tuples = Discrepancy.ExpDesign.bound_tuples
n_sigma2 = len(disc_bound_tuples)
- else:
- n_sigma2 = -len(theta)
- # Check if bias term needs to be inferred
- if Discrepancy.opt_sigma != 'B':
+ # Check if bias term should be inferred
sigma2 = theta[:, -n_sigma2:]
theta = theta[:, :-n_sigma2]
else:
+ n_sigma2 = -len(theta)
sigma2 = None
+
theta = theta if theta.ndim != 1 else theta.reshape((1, -1))
# Evaluate Model/MetaModel at theta
@@ -561,12 +611,11 @@ class MCMC:
"""
BayesObj = self.BayesOpts
- MetaModel = BayesObj.MetaModel
Model = BayesObj.engine.Model
if BayesObj.emulator:
# Evaluate the MetaModel
- mean_pred, std_pred = MetaModel.eval_metamodel(samples=theta)
+ mean_pred, std_pred = BayesObj.engine.MetaModel.eval_metamodel(samples=theta)
else:
# Evaluate the origModel
mean_pred, std_pred = dict(), dict()
@@ -610,8 +659,7 @@ class MCMC:
A error model.
"""
- BayesObj = self.BayesOpts
- MetaModel = BayesObj.MetaModel
+ MetaModel = self.BayesOpts.engine.MetaModel
# Prepare the poster samples
try:
@@ -636,274 +684,6 @@ class MCMC:
# Train a GPR meta-model using MAP
error_MetaModel = MetaModel.create_model_error(
- BayesObj.BiasInputs, y_map, name='Calib')
+ self.BayesOpts.BiasInputs, y_map, name='Calib')
return error_MetaModel
-
- # -------------------------------------------------------------------------
- def gelman_rubin(self, chain, return_var=False):
- """
- The potential scale reduction factor (PSRF) defined by the variance
- within one chain, W, with the variance between chains B.
- Both variances are combined in a weighted sum to obtain an estimate of
- the variance of a parameter \\( \\theta \\).The square root of the
- ratio of this estimates variance to the within chain variance is called
- the potential scale reduction.
- For a well converged chain it should approach 1. Values greater than
- 1.1 typically indicate that the chains have not yet fully converged.
-
- Source: http://joergdietrich.github.io/emcee-convergence.html
-
- https://github.com/jwalton3141/jwalton3141.github.io/blob/master/assets/posts/ESS/rwmh.py
-
- Parameters
- ----------
- chain : array (n_walkers, n_steps, n_params)
- The emcee ensamples.
-
- Returns
- -------
- R_hat : float
- The Gelman-Robin values.
-
- """
- m_chains, n_iters = chain.shape[:2]
-
- # Calculate between-chain variance
- θb = np.mean(chain, axis=1)
- θbb = np.mean(θb, axis=0)
- B_over_n = ((θbb - θb)**2).sum(axis=0)
- B_over_n /= (m_chains - 1)
-
- # Calculate within-chain variances
- ssq = np.var(chain, axis=1, ddof=1)
- W = np.mean(ssq, axis=0)
-
- # (over) estimate of variance
- var_θ = W * (n_iters - 1) / n_iters + B_over_n
-
- if return_var:
- return var_θ
- else:
- # The square root of the ratio of this estimates variance to the
- # within chain variance
- R_hat = np.sqrt(var_θ / W)
- return R_hat
-
- # -------------------------------------------------------------------------
- def marginal_llk_emcee(self, sampler, nburn=None, logp=None, maxiter=1000):
- """
- The Bridge Sampling Estimator of the Marginal Likelihood based on
- https://gist.github.com/junpenglao/4d2669d69ddfe1d788318264cdcf0583
-
- Parameters
- ----------
- sampler : TYPE
- MultiTrace, result of MCMC run.
- nburn : int, optional
- Number of burn-in step. The default is None.
- logp : TYPE, optional
- Model Log-probability function. The default is None.
- maxiter : int, optional
- Maximum number of iterations. The default is 1000.
-
- Returns
- -------
- marg_llk : dict
- Estimated Marginal log-Likelihood.
-
- """
- r0, tol1, tol2 = 0.5, 1e-10, 1e-4
-
- if logp is None:
- logp = sampler.log_prob_fn
-
- # Split the samples into two parts
- # Use the first 50% for fiting the proposal distribution
- # and the second 50% in the iterative scheme.
- if nburn is None:
- mtrace = sampler.chain
- else:
- mtrace = sampler.chain[:, nburn:, :]
-
- nchain, len_trace, nrofVars = mtrace.shape
-
- N1_ = len_trace // 2
- N1 = N1_*nchain
- N2 = len_trace*nchain - N1
-
- samples_4_fit = np.zeros((nrofVars, N1))
- samples_4_iter = np.zeros((nrofVars, N2))
- effective_n = np.zeros((nrofVars))
-
- # matrix with already transformed samples
- for var in range(nrofVars):
-
- # for fitting the proposal
- x = mtrace[:, :N1_, var]
-
- samples_4_fit[var, :] = x.flatten()
- # for the iterative scheme
- x2 = mtrace[:, N1_:, var]
- samples_4_iter[var, :] = x2.flatten()
-
- # effective sample size of samples_4_iter, scalar
- effective_n[var] = self._my_ESS(x2)
-
- # median effective sample size (scalar)
- neff = np.median(effective_n)
-
- # get mean & covariance matrix and generate samples from proposal
- m = np.mean(samples_4_fit, axis=1)
- V = np.cov(samples_4_fit)
- L = chol(V, lower=True)
-
- # Draw N2 samples from the proposal distribution
- gen_samples = m[:, None] + np.dot(
- L, st.norm.rvs(0, 1, size=samples_4_iter.shape)
- )
-
- # Evaluate proposal distribution for posterior & generated samples
- q12 = st.multivariate_normal.logpdf(samples_4_iter.T, m, V)
- q22 = st.multivariate_normal.logpdf(gen_samples.T, m, V)
-
- # Evaluate unnormalized posterior for posterior & generated samples
- q11 = logp(samples_4_iter.T)
- q21 = logp(gen_samples.T)
-
- # Run iterative scheme:
- tmp = self._iterative_scheme(
- N1, N2, q11, q12, q21, q22, r0, neff, tol1, maxiter, 'r'
- )
- if ~np.isfinite(tmp['logml']):
- warnings.warn(
- "Logml could not be estimated within maxiter, rerunning with "
- "adjusted starting value. Estimate might be more variable than"
- " usual.")
- # use geometric mean as starting value
- r0_2 = np.sqrt(tmp['r_vals'][-2]*tmp['r_vals'][-1])
- tmp = self._iterative_scheme(
- q11, q12, q21, q22, r0_2, neff, tol2, maxiter, 'logml'
- )
-
- marg_llk = dict(
- logml=tmp['logml'], niter=tmp['niter'], method="normal",
- q11=q11, q12=q12, q21=q21, q22=q22
- )
- return marg_llk
-
- # -------------------------------------------------------------------------
- def _iterative_scheme(self, N1, N2, q11, q12, q21, q22, r0, neff, tol,
- maxiter, criterion):
- """
- Iterative scheme as proposed in Meng and Wong (1996) to estimate the
- marginal likelihood
-
- """
- l1 = q11 - q12
- l2 = q21 - q22
- # To increase numerical stability,
- # subtracting the median of l1 from l1 & l2 later
- lstar = np.median(l1)
- s1 = neff/(neff + N2)
- s2 = N2/(neff + N2)
- r = r0
- r_vals = [r]
- logml = np.log(r) + lstar
- criterion_val = 1 + tol
-
- i = 0
- while (i <= maxiter) & (criterion_val > tol):
- rold = r
- logmlold = logml
- numi = np.exp(l2 - lstar)/(s1 * np.exp(l2 - lstar) + s2 * r)
- deni = 1/(s1 * np.exp(l1 - lstar) + s2 * r)
- if np.sum(~np.isfinite(numi))+np.sum(~np.isfinite(deni)) > 0:
- warnings.warn(
- """Infinite value in iterative scheme, returning NaN.
- Try rerunning with more samples.""")
- r = (N1/N2) * np.sum(numi)/np.sum(deni)
- r_vals.append(r)
- logml = np.log(r) + lstar
- i += 1
- if criterion == 'r':
- criterion_val = np.abs((r - rold)/r)
- elif criterion == 'logml':
- criterion_val = np.abs((logml - logmlold)/logml)
-
- if i >= maxiter:
- return dict(logml=np.NaN, niter=i, r_vals=np.asarray(r_vals))
- else:
- return dict(logml=logml, niter=i)
-
- # -------------------------------------------------------------------------
- def _my_ESS(self, x):
- """
- Compute the effective sample size of estimand of interest.
- Vectorised implementation.
- https://github.com/jwalton3141/jwalton3141.github.io/blob/master/assets/posts/ESS/rwmh.py
-
-
- Parameters
- ----------
- x : array of shape (n_walkers, n_steps)
- MCMC Samples.
-
- Returns
- -------
- int
- Effective sample size.
-
- """
- m_chains, n_iters = x.shape
-
- def variogram(t):
- variogram = ((x[:, t:] - x[:, :(n_iters - t)])**2).sum()
- variogram /= (m_chains * (n_iters - t))
- return variogram
-
- post_var = self.gelman_rubin(x, return_var=True)
-
- t = 1
- rho = np.ones(n_iters)
- negative_autocorr = False
-
- # Iterate until the sum of consecutive estimates of autocorrelation is
- # negative
- while not negative_autocorr and (t < n_iters):
- rho[t] = 1 - variogram(t) / (2 * post_var)
-
- if not t % 2:
- negative_autocorr = sum(rho[t-1:t+1]) < 0
-
- t += 1
-
- return int(m_chains*n_iters / (1 + 2*rho[1:t].sum()))
-
- # -------------------------------------------------------------------------
- def _check_ranges(self, theta, ranges):
- """
- This function checks if theta lies in the given ranges.
-
- Parameters
- ----------
- theta : array
- Proposed parameter set.
- ranges : nested list
- List of the praremeter ranges.
-
- Returns
- -------
- c : bool
- If it lies in the given range, it return True else False.
-
- """
- c = True
- # traverse in the list1
- for i, bounds in enumerate(ranges):
- x = theta[i]
- # condition check
- if x < bounds[0] or x > bounds[1]:
- c = False
- return c
- return c
diff --git a/src/bayesvalidrox/pylink/pylink.py b/src/bayesvalidrox/pylink/pylink.py
index 227a51ab38cd834e7e85f6193d83563c7ed3437a..637f42317e6f97815e51ce0b331b61c03f26a85b 100644
--- a/src/bayesvalidrox/pylink/pylink.py
+++ b/src/bayesvalidrox/pylink/pylink.py
@@ -231,7 +231,7 @@ class PyLinkForwardModel(object):
self.observations_valid = self.observations_valid
else:
raise Exception("Please provide the observation data as a "
- "dictionary via observations attribute or pass"
+ "dictionary via observations_valid attribute or pass"
" the csv-file path to MeasurementFile "
"attribute")
# Compute the number of observation
diff --git a/src/bayesvalidrox/surrogate_models/__init__.py b/src/bayesvalidrox/surrogate_models/__init__.py
index 70bfb20f570464c2907a0a4128f4ed99b6c13736..6d8ce9f1c58fa154f2af63f0fb5a44097695df33 100644
--- a/src/bayesvalidrox/surrogate_models/__init__.py
+++ b/src/bayesvalidrox/surrogate_models/__init__.py
@@ -1,7 +1,12 @@
# -*- coding: utf-8 -*-
-
+from .engine import Engine
+from .exp_designs import ExpDesigns
+from .input_space import InputSpace
from .surrogate_models import MetaModel
__all__ = [
- "MetaModel"
+ "MetaModel",
+ "InputSpace",
+ "ExpDesigns",
+ "Engine"
]
diff --git a/src/bayesvalidrox/surrogate_models/desktop.ini b/src/bayesvalidrox/surrogate_models/desktop.ini
new file mode 100644
index 0000000000000000000000000000000000000000..632de13ae6b61cecf0d9fdbf9c97cfb16bfb51a4
--- /dev/null
+++ b/src/bayesvalidrox/surrogate_models/desktop.ini
@@ -0,0 +1,2 @@
+[LocalizedFileNames]
+exploration.py=@exploration.py,0
diff --git a/src/bayesvalidrox/surrogate_models/engine.py b/src/bayesvalidrox/surrogate_models/engine.py
index 42307d4770d4ae23a40107dfea64057aac682c23..1c8fa56e6acf84fd53cabf92843e04dcce4835c0 100644
--- a/src/bayesvalidrox/surrogate_models/engine.py
+++ b/src/bayesvalidrox/surrogate_models/engine.py
@@ -3,35 +3,27 @@
Engine to train the surrogate
"""
-import copy
from copy import deepcopy, copy
-import h5py
import joblib
+from joblib import Parallel, delayed
+import matplotlib.pyplot as plt
+import multiprocessing
import numpy as np
import os
-
+import pandas as pd
+import pathlib
+import scipy.optimize as opt
from scipy import stats, signal, linalg, sparse
from scipy.spatial import distance
-from tqdm import tqdm
-import scipy.optimize as opt
from sklearn.metrics import mean_squared_error
-import multiprocessing
-import matplotlib.pyplot as plt
-import pandas as pd
-import sys
import seaborn as sns
-from joblib import Parallel, delayed
-
+import sys
+from tqdm import tqdm
from bayesvalidrox.bayes_inference.bayes_inference import BayesInference
from bayesvalidrox.bayes_inference.discrepancy import Discrepancy
from .exploration import Exploration
-import pathlib
-#from .inputs import Input
-#from .exp_designs import ExpDesigns
-#from .surrogate_models import MetaModel
-#from bayesvalidrox.post_processing.post_processing import PostProcessing
def hellinger_distance(P, Q):
"""
@@ -55,17 +47,17 @@ def hellinger_distance(P, Q):
"""
P = np.array(P)
- Q= np.array(Q)
-
+ Q = np.array(Q)
+
mu1 = P.mean()
Sigma1 = np.std(P)
mu2 = Q.mean()
Sigma2 = np.std(Q)
- term1 = np.sqrt(2*Sigma1*Sigma2 / (Sigma1**2 + Sigma2**2))
+ term1 = np.sqrt(2 * Sigma1 * Sigma2 / (Sigma1 ** 2 + Sigma2 ** 2))
- term2 = np.exp(-.25 * (mu1 - mu2)**2 / (Sigma1**2 + Sigma2**2))
+ term2 = np.exp(-.25 * (mu1 - mu2) ** 2 / (Sigma1 ** 2 + Sigma2 ** 2))
H_squared = 1 - term1 * term2
@@ -100,9 +92,10 @@ def logpdf(x, mean, cov):
return log_lik
+
def subdomain(Bounds, n_new_samples):
"""
- Divides a domain defined by Bounds into sub domains.
+ Divides a domain defined by Bounds into subdomains.
Parameters
----------
@@ -110,8 +103,6 @@ def subdomain(Bounds, n_new_samples):
List of lower and upper bounds.
n_new_samples : int
Number of samples to divide the domain for.
- n_params : int
- The number of params to build the subdomains for
Returns
-------
@@ -127,23 +118,41 @@ def subdomain(Bounds, n_new_samples):
LinSpace[i] = np.linspace(start=Bounds[i][0], stop=Bounds[i][1],
num=n_subdomains)
Subdomains = []
- for k in range(n_subdomains-1):
+ for k in range(n_subdomains - 1):
mylist = []
for i in range(n_params):
- mylist.append((LinSpace[i, k+0], LinSpace[i, k+1]))
+ mylist.append((LinSpace[i, k + 0], LinSpace[i, k + 1]))
Subdomains.append(tuple(mylist))
return Subdomains
-class Engine():
-
-
+
+class Engine:
+
def __init__(self, MetaMod, Model, ExpDes):
self.MetaModel = MetaMod
self.Model = Model
self.ExpDesign = ExpDes
self.parallel = False
-
+ self.trained = False
+
+ # Init other parameters
+ self.bound_tuples = None
+ self.errorModel = None
+ self.LCerror = None
+ self.n_obs = None
+ self.observations = None
+ self.out_names = None
+ self.seqMinDist = None
+ self.seqRMSEStd = None
+ self.SeqKLD = None
+ self.SeqDistHellinger = None
+ self.SeqBME = None
+ self.seqValidError = None
+ self.SeqModifiedLOO = None
+ self.valid_likelihoods = None
+ self._y_hat_prev = None
+
def start_engine(self) -> None:
"""
Do all the preparations that need to be run before the actual training
@@ -155,9 +164,8 @@ class Engine():
"""
self.out_names = self.Model.Output.names
self.MetaModel.out_names = self.out_names
-
-
- def train_normal(self, parallel = False, verbose = False, save = False) -> None:
+
+ def train_normal(self, parallel=False, verbose=False, save=False) -> None:
"""
Trains surrogate on static samples only.
Samples are taken from the experimental design and the specified
@@ -170,10 +178,12 @@ class Engine():
None
"""
-
+ if self.out_names == 'None':
+ self.start_engine()
+
ExpDesign = self.ExpDesign
MetaModel = self.MetaModel
-
+
# Read ExpDesign (training and targets) from the provided hdf5
if ExpDesign.hdf5_file is not None:
# TODO: need to run 'generate_ED' as well after this or not?
@@ -182,51 +192,52 @@ class Engine():
# Check if an old hdf5 file exists: if yes, rename it
hdf5file = f'ExpDesign_{self.Model.name}.hdf5'
if os.path.exists(hdf5file):
- # os.rename(hdf5file, 'old_'+hdf5file)
+ # os.rename(hdf5file, 'old_'+hdf5file)
file = pathlib.Path(hdf5file)
file.unlink()
# Prepare X samples
# For training the surrogate use ExpDesign.X_tr, ExpDesign.X is for the model to run on
ExpDesign.generate_ED(ExpDesign.n_init_samples,
- transform=True,
- max_pce_deg=np.max(MetaModel.pce_deg))
-
+ transform=True,
+ max_pce_deg=np.max(MetaModel.pce_deg))
+
# Run simulations at X
if not hasattr(ExpDesign, 'Y') or ExpDesign.Y is None:
print('\n Now the forward model needs to be run!\n')
- ED_Y, up_ED_X = self.Model.run_model_parallel(ExpDesign.X, mp = parallel)
+ ED_Y, up_ED_X = self.Model.run_model_parallel(ExpDesign.X, mp=parallel)
ExpDesign.Y = ED_Y
else:
# Check if a dict has been passed.
if not type(ExpDesign.Y) is dict:
raise Exception('Please provide either a dictionary or a hdf5'
'file to ExpDesign.hdf5_file argument.')
-
+
# Separate output dict and x-values
if 'x_values' in ExpDesign.Y:
ExpDesign.x_values = ExpDesign.Y['x_values']
del ExpDesign.Y['x_values']
else:
print('No x_values are given, this might lead to issues during PostProcessing')
-
-
+
# Fit the surrogate
MetaModel.fit(ExpDesign.X, ExpDesign.Y, parallel, verbose)
-
+
# Save what there is to save
if save:
# Save surrogate
with open(f'surrogates/surrogate_{self.Model.name}.pk1', 'wb') as output:
joblib.dump(MetaModel, output, 2)
-
+
# Zip the model run directories
- if self.Model.link_type.lower() == 'pylink' and\
- self.ExpDesign.sampling_method.lower() != 'user':
+ if self.Model.link_type.lower() == 'pylink' and \
+ self.ExpDesign.sampling_method.lower() != 'user':
self.Model.zip_subdirs(self.Model.name, f'{self.Model.name}_')
-
-
- def train_sequential(self, parallel = False, verbose = False) -> None:
+
+ # Set that training was done
+ self.trained = True
+
+ def train_sequential(self, parallel=False, verbose=False) -> None:
"""
Train the surrogate in a sequential manner.
First build and train evereything on the static samples, then iterate
@@ -237,22 +248,21 @@ class Engine():
None
"""
- #self.train_normal(parallel, verbose)
+ # self.train_normal(parallel, verbose)
self.parallel = parallel
self.train_seq_design(parallel, verbose)
-
-
+
# -------------------------------------------------------------------------
def eval_metamodel(self, samples=None, nsamples=None,
sampling_method='random', return_samples=False):
"""
- Evaluates meta-model at the requested samples. One can also generate
+ Evaluates metamodel at the requested samples. One can also generate
nsamples.
Parameters
----------
samples : array of shape (n_samples, n_params), optional
- Samples to evaluate meta-model at. The default is None.
+ Samples to evaluate metamodel at. The default is None.
nsamples : int, optional
Number of samples to generate, if no `samples` is provided. The
default is None.
@@ -275,7 +285,7 @@ class Engine():
samples = self.ExpDesign.generate_samples(
nsamples,
sampling_method
- )
+ )
# Transformation to other space is to be done in the MetaModel
# TODO: sort the transformations better
@@ -285,10 +295,9 @@ class Engine():
return mean_pred, std_pred, samples
else:
return mean_pred, std_pred
-
-
+
# -------------------------------------------------------------------------
- def train_seq_design(self, parallel = False, verbose = False):
+ def train_seq_design(self, parallel=False, verbose=False):
"""
Starts the adaptive sequential design for refining the surrogate model
by selecting training points in a sequential manner.
@@ -300,7 +309,7 @@ class Engine():
"""
self.parallel = parallel
-
+
# Initialization
self.SeqModifiedLOO = {}
self.seqValidError = {}
@@ -310,14 +319,13 @@ class Engine():
self.seqRMSEMean = {}
self.seqRMSEStd = {}
self.seqMinDist = []
-
- if not hasattr(self.MetaModel, 'valid_samples'):
+
+ if not hasattr(self.MetaModel, 'valid_samples') or self.MetaModel.valid_samples is None:
self.ExpDesign.valid_samples = []
self.ExpDesign.valid_model_runs = []
self.valid_likelihoods = []
-
- validError = None
+ # validError = None
# Determine the metamodel type
if self.MetaModel.meta_model_type.lower() != 'gpe':
@@ -336,14 +344,15 @@ class Engine():
n_replication = self.ExpDesign.n_replication
util_func = self.ExpDesign.util_func
output_name = self.out_names
-
+
# Handle if only one UtilityFunctions is provided
if not isinstance(util_func, list):
util_func = [self.ExpDesign.util_func]
# Read observations or MCReference
# TODO: recheck the logic in this if statement
- if (len(self.Model.observations) != 0 or self.Model.meas_file is not None) and hasattr(self.MetaModel, 'Discrepancy'):
+ if (len(self.Model.observations) != 0 or self.Model.meas_file is not None) and hasattr(self.MetaModel,
+ 'Discrepancy'):
self.observations = self.Model.read_observation()
obs_data = self.observations
else:
@@ -351,15 +360,16 @@ class Engine():
# TODO: TotalSigma2 not defined if not in this else???
# TODO: no self.observations if in here
TotalSigma2 = {}
-
+
# ---------- Initial self.MetaModel ----------
- self.train_normal(parallel = parallel, verbose=verbose)
-
+ if not self.trained:
+ self.train_normal(parallel=parallel, verbose=verbose)
+
initMetaModel = deepcopy(self.MetaModel)
# Validation error if validation set is provided.
if self.ExpDesign.valid_model_runs:
- init_rmse, init_valid_error = self._validError(initMetaModel)
+ init_rmse, init_valid_error = self._validError() # initMetaModel)
init_valid_error = list(init_valid_error.values())
else:
init_rmse = None
@@ -379,7 +389,7 @@ class Engine():
if post_snapshot:
parNames = self.ExpDesign.par_names
print('Posterior snapshot (initial) is being plotted...')
- self.__posteriorPlot(init_post, parNames, 'SeqPosterior_init')
+ self._posteriorPlot(init_post, parNames, 'SeqPosterior_init')
# Check the convergence of the Mean & Std
if mc_ref and pce:
@@ -390,13 +400,14 @@ class Engine():
# Read the initial experimental design
Xinit = self.ExpDesign.X
init_n_samples = len(self.ExpDesign.X)
- initYprev = self.ExpDesign.Y#initMetaModel.ModelOutputDict
- #self.MetaModel.ModelOutputDict = self.ExpDesign.Y
+ initYprev = self.ExpDesign.Y # initMetaModel.ModelOutputDict
+ # self.MetaModel.ModelOutputDict = self.ExpDesign.Y
initLCerror = initMetaModel.LCerror
n_itrs = max_n_samples - init_n_samples
- ## Get some initial statistics
+ # Get some initial statistics
# Read the initial ModifiedLOO
+ init_mod_LOO = []
if pce:
Scores_all, varExpDesignY = [], []
for out_name in output_name:
@@ -412,14 +423,14 @@ class Engine():
Scores = [item for sublist in Scores_all for item in sublist]
weights = [item for sublist in varExpDesignY for item in sublist]
- init_mod_LOO = [np.average([1-score for score in Scores],
+ init_mod_LOO = [np.average([1 - score for score in Scores],
weights=weights)]
prevMetaModel_dict = {}
- #prevExpDesign_dict = {}
+ # prevExpDesign_dict = {}
# Can run sequential design multiple times for comparison
for repIdx in range(n_replication):
- print(f'\n>>>> Replication: {repIdx+1}<<<<')
+ print(f'\n>>>> Replication: {repIdx + 1}<<<<')
# util_func: the function to use inside the type of exploitation
for util_f in util_func:
@@ -435,7 +446,6 @@ class Engine():
Yprev = initYprev
Xfull = []
- Yfull = []
# Store the initial ModifiedLOO
if pce:
@@ -457,23 +467,23 @@ class Engine():
# ------- Start Sequential Experimental Design -------
postcnt = 1
- for itr_no in range(1, n_itrs+1):
+ for itr_no in range(1, n_itrs + 1):
print(f'\n>>>> Iteration number {itr_no} <<<<')
# Save the metamodel prediction before updating
prevMetaModel_dict[itr_no] = deepcopy(self.MetaModel)
- #prevExpDesign_dict[itr_no] = deepcopy(self.ExpDesign)
+ # prevExpDesign_dict[itr_no] = deepcopy(self.ExpDesign)
if itr_no > 1:
- pc_model = prevMetaModel_dict[itr_no-1]
+ pc_model = prevMetaModel_dict[itr_no - 1]
self._y_hat_prev, _ = pc_model.eval_metamodel(
samples=Xfull[-1].reshape(1, -1))
- del prevMetaModel_dict[itr_no-1]
+ del prevMetaModel_dict[itr_no - 1]
# Optimal Bayesian Design
- #self.MetaModel.ExpDesignFlag = 'sequential'
+ # self.MetaModel.ExpDesignFlag = 'sequential'
Xnew, updatedPrior = self.choose_next_sample(TotalSigma2,
- n_canddidate,
- util_f)
+ n_canddidate,
+ util_f)
S = np.min(distance.cdist(Xinit, Xnew, 'euclidean'))
self.seqMinDist.append(S)
print(f"\nmin Dist from OldExpDesign: {S:2f}")
@@ -482,20 +492,19 @@ class Engine():
# Evaluate the full model response at the new sample
Ynew, _ = self.Model.run_model_parallel(
Xnew, prevRun_No=total_n_samples
- )
+ )
total_n_samples += Xnew.shape[0]
# ------ Plot the surrogate model vs Origninal Model ------
- if hasattr(self.ExpDesign, 'adapt_verbose') and \
- self.ExpDesign.adapt_verbose:
+ if self.ExpDesign.adapt_verbose:
from .adaptPlot import adaptPlot
y_hat, std_hat = self.MetaModel.eval_metamodel(
samples=Xnew
- )
+ )
adaptPlot(
self.MetaModel, Ynew, y_hat, std_hat,
plotED=False
- )
+ )
# -------- Retrain the surrogate model -------
# Extend new experimental design
@@ -509,11 +518,11 @@ class Engine():
# Pass new design to the metamodel object
self.ExpDesign.sampling_method = 'user'
self.ExpDesign.X = Xfull
- #self.ExpDesign.Y = self.MetaModel.ModelOutputDict
+ # self.ExpDesign.Y = self.MetaModel.ModelOutputDict
# Save the Experimental Design for next iteration
Xprev = Xfull
- Yprev = self.ExpDesign.Y
+ Yprev = self.ExpDesign.Y
# Pass the new prior as the input
# TODO: another look at this - no difference apc to pce to gpe?
@@ -550,7 +559,7 @@ class Engine():
weights = [item for sublist in varExpDesignY for item
in sublist]
ModifiedLOO = [np.average(
- [1-score for score in Scores], weights=weights)]
+ [1 - score for score in Scores], weights=weights)]
print('\n')
print(f"Updated ModifiedLOO {util_f}:\n", ModifiedLOO)
@@ -558,7 +567,7 @@ class Engine():
# Compute the validation error
if self.ExpDesign.valid_model_runs:
- rmse, validError = self._validError(self.MetaModel)
+ rmse, validError = self._validError() # self.MetaModel)
ValidError = list(validError.values())
else:
rmse = None
@@ -586,8 +595,8 @@ class Engine():
if post_snapshot and postcnt % step_snapshot == 0:
parNames = self.ExpDesign.par_names
print('Posterior snapshot is being plotted...')
- self.__posteriorPlot(Posterior, parNames,
- f'SeqPosterior_{postcnt}')
+ self._posteriorPlot(Posterior, parNames,
+ f'SeqPosterior_{postcnt}')
postcnt += 1
# Check the convergence of the Mean&Std
@@ -617,11 +626,11 @@ class Engine():
if len(obs_data) != 0:
del out
print()
- print('-'*50)
+ print('-' * 50)
print()
# Store updated ModifiedLOO and BME in dictonary
- strKey = f'{util_f}_rep_{repIdx+1}'
+ strKey = f'{util_f}_rep_{repIdx + 1}'
if pce:
self.SeqModifiedLOO[strKey] = SeqModifiedLOO
if len(self.ExpDesign.valid_model_runs) != 0:
@@ -632,7 +641,7 @@ class Engine():
self.SeqBME[strKey] = SeqBME
self.SeqKLD[strKey] = SeqKLD
if hasattr(self.MetaModel, 'valid_likelihoods') and \
- self.valid_likelihoods:
+ self.valid_likelihoods:
self.SeqDistHellinger[strKey] = SeqDistHellinger
if mc_ref and pce:
self.seqRMSEMean[strKey] = seqRMSEMean
@@ -655,7 +664,7 @@ class Engine():
Candidate samples.
index : int
Model output index.
- UtilMethod : string, optional
+ util_func : string, optional
Exploitation utility function. The default is 'Entropy'.
Returns
@@ -673,10 +682,11 @@ class Engine():
X_can = X_can.reshape(1, -1)
Y_PC_can, std_PC_can = MetaModel.eval_metamodel(samples=X_can)
+ score = None
if util_func.lower() == 'alm':
# ----- Entropy/MMSE/active learning MacKay(ALM) -----
# Compute perdiction variance of the old model
- canPredVar = {key: std_PC_can[key]**2 for key in out_names}
+ canPredVar = {key: std_PC_can[key] ** 2 for key in out_names}
varPCE = np.zeros((len(out_names), X_can.shape[0]))
for KeyIdx, key in enumerate(out_names):
@@ -691,7 +701,7 @@ class Engine():
# Compute perdiction error and variance of the old model
predError = {key: Y_PC_can[key] for key in out_names}
- canPredVar = {key: std_PC_can[key]**2 for key in out_names}
+ canPredVar = {key: std_PC_can[key] ** 2 for key in out_names}
# Compute perdiction error and variance of the old model
# Eq (5) from Liu et al.(2018)
@@ -699,10 +709,10 @@ class Engine():
for KeyIdx, key in enumerate(out_names):
residual = predError[key] - out_dict_y[key][int(index)]
var = canPredVar[key]
- EIGF_PCE[KeyIdx] = np.max(residual**2 + var, axis=1)
+ EIGF_PCE[KeyIdx] = np.max(residual ** 2 + var, axis=1)
score = np.max(EIGF_PCE, axis=0)
- return -1 * score # -1 is for minimization instead of maximization
+ return -1 * score # -1 is for minimization instead of maximization
# -------------------------------------------------------------------------
def util_BayesianActiveDesign(self, y_hat, std, sigma2Dict, var='DKL'):
@@ -716,8 +726,8 @@ class Engine():
Parameters
----------
- X_can : array of shape (n_samples, n_params)
- Candidate samples.
+ y_hat : unknown
+ std : unknown
sigma2Dict : dict
A dictionary containing the measurement errors (sigma^2).
var : string, optional
@@ -742,14 +752,12 @@ class Engine():
# Sample a distribution for a normal dist
# with Y_mean_can as the mean and Y_std_can as std.
Y_MC, std_MC = {}, {}
- logPriorLikelihoods = np.zeros((mc_size))
- # print(y_hat)
- # print(list[y_hat])
+ logPriorLikelihoods = np.zeros(mc_size)
for key in list(y_hat):
- cov = np.diag(std[key]**2)
- # print(y_hat[key], cov)
+ cov = np.diag(std[key] ** 2)
+ print(key, y_hat[key], std[key])
# TODO: added the allow_singular = True here
- rv = stats.multivariate_normal(mean=y_hat[key], cov=cov,)
+ rv = stats.multivariate_normal(mean=y_hat[key], cov=cov, allow_singular=True)
Y_MC[key] = rv.rvs(size=mc_size)
logPriorLikelihoods += rv.logpdf(Y_MC[key])
std_MC[key] = np.zeros((mc_size, y_hat[key].shape[0]))
@@ -757,16 +765,16 @@ class Engine():
# Likelihood computation (Comparison of data and simulation
# results via PCE with candidate design)
likelihoods = self._normpdf(Y_MC, std_MC, obs_data, sigma2Dict)
-
+
# Rejection Step
# Random numbers between 0 and 1
unif = np.random.rand(1, mc_size)[0]
# Reject the poorly performed prior
- accepted = (likelihoods/np.max(likelihoods)) >= unif
+ accepted = (likelihoods / np.max(likelihoods)) >= unif
# Prior-based estimation of BME
- logBME = np.log(np.nanmean(likelihoods), dtype=np.longdouble)#float128)
+ logBME = np.log(np.nanmean(likelihoods), dtype=np.longdouble) # float128)
# Posterior-based expectation of likelihoods
postLikelihoods = likelihoods[accepted]
@@ -778,6 +786,7 @@ class Engine():
# Utility function Eq.2 in Ref. (2)
# Posterior covariance matrix after observing data y
# Kullback-Leibler Divergence (Sergey's paper)
+ U_J_d = None
if var == 'DKL':
# TODO: Calculate the correction factor for BME
@@ -813,7 +822,7 @@ class Engine():
AIC = -2 * maxlogL + 2 * nModelParams
# 2 * nModelParams * (nModelParams+1) / (n_obs-nModelParams-1)
penTerm = 0
- U_J_d = 1*(AIC + penTerm)
+ U_J_d = 1 * (AIC + penTerm)
# Deviance information criterion
elif var == 'DIC':
@@ -821,7 +830,7 @@ class Engine():
N_star_p = 0.5 * np.var(np.log(likelihoods[likelihoods != 0]))
Likelihoods_theta_mean = self._normpdf(
y_hat, std, obs_data, sigma2Dict
- )
+ )
DIC = -2 * np.log(Likelihoods_theta_mean) + 2 * N_star_p
U_J_d = DIC
@@ -838,7 +847,7 @@ class Engine():
del Y_MC
del std_MC
- return -1 * U_J_d # -1 is for minimization instead of maximization
+ return -1 * U_J_d # -1 is for minimization instead of maximization
# -------------------------------------------------------------------------
def util_BayesianDesign(self, X_can, X_MC, sigma2Dict, var='DKL'):
@@ -849,6 +858,7 @@ class Engine():
----------
X_can : array of shape (n_samples, n_params)
Candidate samples.
+ X_MC : unknown
sigma2Dict : dict
A dictionary containing the measurement errors (sigma^2).
var : string, optional
@@ -887,11 +897,11 @@ class Engine():
for key in oldExpDesignY.keys():
NewExpDesignY[key] = np.vstack(
(oldExpDesignY[key], Y_PC_can[key])
- )
+ )
engine_can.ExpDesign.sampling_method = 'user'
engine_can.ExpDesign.X = NewExpDesignX
- #engine_can.ModelOutputDict = NewExpDesignY
+ # engine_can.ModelOutputDict = NewExpDesignY
engine_can.ExpDesign.Y = NewExpDesignY
# Train the model for the observed data using x_can
@@ -899,7 +909,7 @@ class Engine():
engine_can.start_engine()
engine_can.train_normal(parallel=False)
engine_can.MetaModel.fit(NewExpDesignX, NewExpDesignY)
-# engine_can.train_norm_design(parallel=False)
+ # engine_can.train_norm_design(parallel=False)
# Set the ExpDesign to its original values
engine_can.ExpDesign.X = oldExpDesignX
@@ -907,7 +917,7 @@ class Engine():
engine_can.ExpDesign.Y = oldExpDesignY
if var.lower() == 'mi':
- # Mutual information based on Krause et al
+ # Mutual information based on Krause et al.
# Adapted from Beck & Guillas (MICE) paper
_, std_PC_can = engine_can.MetaModel.eval_metamodel(samples=X_can)
std_can = {key: std_PC_can[key] for key in out_names}
@@ -916,7 +926,7 @@ class Engine():
varPCE = np.zeros((len(out_names)))
for i, key in enumerate(out_names):
- varPCE[i] = np.mean(std_old[key]**2/std_can[key]**2)
+ varPCE[i] = np.mean(std_old[key] ** 2 / std_can[key] ** 2)
score = np.mean(varPCE)
return -1 * score
@@ -932,9 +942,9 @@ class Engine():
# Compute the score
score = []
for i, key in enumerate(out_names):
- pce_var = Y_MC_std_can[key]**2
- pce_var_can = Y_MC_std[key]**2
- score.append(np.mean(pce_var-pce_var_can, axis=0))
+ pce_var = Y_MC_std_can[key] ** 2
+ pce_var_can = Y_MC_std[key] ** 2
+ score.append(np.mean(pce_var - pce_var_can, axis=0))
score = np.mean(score)
return -1 * score
@@ -944,13 +954,14 @@ class Engine():
MCsize = X_MC.shape[0]
ESS = 0
- while ((ESS > MCsize) or (ESS < 1)):
+ likelihoods = None
+ while (ESS > MCsize) or (ESS < 1):
# Enriching Monte Carlo samples if need be
if ESS != 0:
X_MC = self.ExpDesign.generate_samples(
MCsize, 'random'
- )
+ )
# Evaluate the MetaModel at the given samples
Y_MC, std_MC = PCE_Model_can.eval_metamodel(samples=X_MC)
@@ -959,13 +970,13 @@ class Engine():
# results via PCE with candidate design)
likelihoods = self._normpdf(
Y_MC, std_MC, self.observations, sigma2Dict
- )
+ )
# Check the Effective Sample Size (1<ESS<MCsize)
- ESS = 1 / np.sum(np.square(likelihoods/np.sum(likelihoods)))
+ ESS = 1 / np.sum(np.square(likelihoods / np.sum(likelihoods)))
# Enlarge sample size if it doesn't fulfill the criteria
- if ((ESS > MCsize) or (ESS < 1)):
+ if (ESS > MCsize) or (ESS < 1):
print("--- increasing MC size---")
MCsize *= 10
ESS = 0
@@ -975,19 +986,20 @@ class Engine():
unif = np.random.rand(1, MCsize)[0]
# Reject the poorly performed prior
- accepted = (likelihoods/np.max(likelihoods)) >= unif
+ accepted = (likelihoods / np.max(likelihoods)) >= unif
# -------------------- Utility functions --------------------
# Utility function Eq.2 in Ref. (2)
# Kullback-Leibler Divergence (Sergey's paper)
+ U_J_d = None
if var == 'DKL':
# Prior-based estimation of BME
- logBME = np.log(np.nanmean(likelihoods, dtype=np.longdouble))#float128))
+ logBME = np.log(np.nanmean(likelihoods, dtype=np.longdouble)) # float128))
# Posterior-based expectation of likelihoods
- postLikelihoods = likelihoods[accepted]
- postExpLikelihoods = np.mean(np.log(postLikelihoods))
+ # postLikelihoods = likelihoods[accepted]
+ # postExpLikelihoods = np.mean(np.log(postLikelihoods))
# Haun et al implementation
U_J_d = np.mean(np.log(likelihoods[likelihoods != 0]) - logBME)
@@ -1022,6 +1034,8 @@ class Engine():
postExpLikelihoods = np.mean(np.log(postLikelihoods))
# Posterior-based expectation of prior densities
+ logPriorLikelihoods = []
+ logPriorLikelihoods[accepted] = None # TODO: this is not defined here, just a fix
postExpPrior = np.mean(logPriorLikelihoods[accepted])
infEntropy = logBME - postExpPrior - postExpLikelihoods
@@ -1048,8 +1062,7 @@ class Engine():
del Y_MC
del std_MC
- return -1 * U_J_d # -1 is for minimization instead of maximization
-
+ return -1 * U_J_d # -1 is for minimization instead of maximization
# -------------------------------------------------------------------------
def run_util_func(self, method, candidates, index, sigma2Dict=None,
@@ -1092,7 +1105,7 @@ class Engine():
elif method.lower() == 'bayesactdesign':
NCandidate = candidates.shape[0]
- U_J_d = np.zeros((NCandidate))
+ U_J_d = np.zeros(NCandidate)
# Evaluate all candidates
y_can, std_can = self.MetaModel.eval_metamodel(samples=candidates)
# loop through candidates
@@ -1100,20 +1113,20 @@ class Engine():
desc="BAL Design"):
y_hat = {key: items[idx] for key, items in y_can.items()}
std = {key: items[idx] for key, items in std_can.items()}
-
- # print(y_hat)
- # print(std)
+
+ # print(y_hat)
+ # print(std)
U_J_d[idx] = self.util_BayesianActiveDesign(
y_hat, std, sigma2Dict, var)
elif method.lower() == 'bayesoptdesign':
NCandidate = candidates.shape[0]
- U_J_d = np.zeros((NCandidate))
+ U_J_d = np.zeros(NCandidate)
for idx, X_can in tqdm(enumerate(candidates), ascii=True,
desc="OptBayesianDesign"):
U_J_d[idx] = self.util_BayesianDesign(X_can, X_MC, sigma2Dict,
var)
- return (index, -1 * U_J_d)
+ return index, -1 * U_J_d
# -------------------------------------------------------------------------
def dual_annealing(self, method, Bounds, sigma2Dict, var, Run_No,
@@ -1130,6 +1143,7 @@ class Engine():
List of lower and upper boundaries of parameters.
sigma2Dict : dict
A dictionary containing the measurement errors (sigma^2).
+ var : unknown
Run_No : int
Run number.
verbose : bool, optional
@@ -1147,13 +1161,14 @@ class Engine():
Model = self.Model
max_func_itr = self.ExpDesign.max_func_itr
- if method == 'VarOptDesign':
+ Res_Global = None
+ if method.lower() == 'varoptdesign':
Res_Global = opt.dual_annealing(self.util_VarBasedDesign,
bounds=Bounds,
args=(Model, var),
maxfun=max_func_itr)
- elif method == 'BayesOptDesign':
+ elif method.lower() == 'bayesoptdesign':
Res_Global = opt.dual_annealing(self.util_BayesianDesign,
bounds=Bounds,
args=(Model, sigma2Dict, var),
@@ -1163,7 +1178,7 @@ class Engine():
print(f"Global minimum: xmin = {Res_Global.x}, "
f"f(xmin) = {Res_Global.fun:.6f}, nfev = {Res_Global.nfev}")
- return (Run_No, Res_Global.x)
+ return Run_No, Res_Global.x
# -------------------------------------------------------------------------
def tradeoff_weights(self, tradeoff_scheme, old_EDX, old_EDY):
@@ -1174,7 +1189,7 @@ class Engine():
`None`: No exploration.
`equal`: Same weights for exploration and exploitation scores.
`epsilon-decreasing`: Start with more exploration and increase the
- influence of exploitation along the way with a exponential decay
+ influence of exploitation along the way with an exponential decay
function
`adaptive`: An adaptive method based on:
Liu, Haitao, Jianfei Cai, and Yew-Soon Ong. "An adaptive sampling
@@ -1198,6 +1213,8 @@ class Engine():
Exploitation weight.
"""
+ exploration_weight = None
+
if tradeoff_scheme is None:
exploration_weight = 0
@@ -1207,22 +1224,22 @@ class Engine():
elif tradeoff_scheme == 'epsilon-decreasing':
# epsilon-decreasing scheme
# Start with more exploration and increase the influence of
- # exploitation along the way with a exponential decay function
+ # exploitation along the way with an exponential decay function
initNSamples = self.ExpDesign.n_init_samples
n_max_samples = self.ExpDesign.n_max_samples
itrNumber = (self.ExpDesign.X.shape[0] - initNSamples)
itrNumber //= self.ExpDesign.n_new_samples
- tau2 = -(n_max_samples-initNSamples-1) / np.log(1e-8)
- exploration_weight = signal.exponential(n_max_samples-initNSamples,
+ tau2 = -(n_max_samples - initNSamples - 1) / np.log(1e-8)
+ exploration_weight = signal.exponential(n_max_samples - initNSamples,
0, tau2, False)[itrNumber]
elif tradeoff_scheme == 'adaptive':
# Extract itrNumber
initNSamples = self.ExpDesign.n_init_samples
- n_max_samples = self.ExpDesign.n_max_samples
+ # n_max_samples = self.ExpDesign.n_max_samples
itrNumber = (self.ExpDesign.X.shape[0] - initNSamples)
itrNumber //= self.ExpDesign.n_new_samples
@@ -1241,7 +1258,7 @@ class Engine():
pce_y_prev = np.array(list(self._y_hat_prev.values()))[:, 0]
mseCVError = mean_squared_error(pce_y_prev, y)
- exploration_weight = min([0.5*mseError/mseCVError, 1])
+ exploration_weight = min([0.5 * mseError / mseCVError, 1])
# Exploitation weight
exploitation_weight = 1 - exploration_weight
@@ -1292,8 +1309,11 @@ class Engine():
# -----------------------------------------
# Utility function exploit_method provided by user
if exploit_method.lower() == 'user':
- if not hasattr(self.ExpDesign, 'ExploitFunction'):
- raise AttributeError('Function `ExploitFunction` not given to the ExpDesign, thus cannor run user-defined sequential scheme')
+ # TODO: is the exploit_method meant here?
+ if not hasattr(self.ExpDesign, 'ExploitFunction') or self.ExpDesign.ExploitFunction is None:
+ raise AttributeError(
+ 'Function `ExploitFunction` not given to the ExpDesign, thus cannor run user-defined sequential'
+ 'scheme')
# TODO: syntax does not fully match the rest - can test this??
Xnew, filteredSamples = self.ExpDesign.ExploitFunction(self)
@@ -1302,7 +1322,6 @@ class Engine():
return Xnew, filteredSamples
-
# Dual-Annealing works differently from the rest, so deal with this first
# Here exploration and exploitation are performed simulataneously
if explore_method == 'dual annealing':
@@ -1330,7 +1349,7 @@ class Engine():
results = []
for i in range(n_new_samples):
results.append(self.dual_annealing(exploit_method, subdomains[i], sigma2, var, i))
-
+
# New sample
Xnew = np.array([results[i][1] for i in range(n_new_samples)])
print("\nXnew:\n", Xnew)
@@ -1338,29 +1357,30 @@ class Engine():
# Computational cost
elapsed_time = time.time() - start_time
print("\n")
- print(f"Elapsed_time: {round(elapsed_time,2)} sec.")
- print('-'*20)
-
+ print(f"Elapsed_time: {round(elapsed_time, 2)} sec.")
+ print('-' * 20)
+
return Xnew, None
-
+
# Generate needed Exploration class
explore = Exploration(self.ExpDesign, n_candidates)
explore.w = 100 # * ndim #500 # TODO: where does this value come from?
-
+
# Select criterion (mc-intersite-proj-th, mc-intersite-proj)
explore.mc_criterion = 'mc-intersite-proj'
-
+
# Generate the candidate samples
# TODO: here use the sampling method provided by the expdesign?
- sampling_method = self.ExpDesign.sampling_method
-
+ # sampling_method = self.ExpDesign.sampling_method
+
# TODO: changed this from 'random' for LOOCV
- if explore_method == 'LOOCV':
- allCandidates = self.ExpDesign.generate_samples(n_candidates,
- sampling_method)
- else:
- allCandidates, scoreExploration = explore.get_exploration_samples()
-
+ # TODO: these are commented out as they are not used !?
+ # if explore_method == 'LOOCV':
+ # allCandidates = self.ExpDesign.generate_samples(n_candidates,
+ # sampling_method)
+ # else:
+ # allCandidates, scoreExploration = explore.get_exploration_samples()
+
# -----------------------------------------
# ---------- EXPLORATION METHODS ----------
# -----------------------------------------
@@ -1372,7 +1392,7 @@ class Engine():
# Generate random samples
allCandidates = self.ExpDesign.generate_samples(n_candidates,
- 'random')
+ 'random')
# Construct error model based on LCerror
errorModel = self.MetaModel.create_ModelError(old_EDX, self.LCerror)
@@ -1400,6 +1420,20 @@ class Engine():
if ndim == 2:
def plotter(points, allCandidates, Method,
scoreExploration=None):
+ """
+ unknown
+
+ Parameters
+ ----------
+ points
+ allCandidates
+ Method
+ scoreExploration
+
+ Returns
+ -------
+
+ """
if Method == 'Voronoi':
from scipy.spatial import Voronoi, voronoi_plot_2d
vor = Voronoi(points)
@@ -1413,7 +1447,7 @@ class Engine():
ax1.scatter(allCandidates[:, 0], allCandidates[:, 1], s=10,
c='b', marker="o", label='Design candidates')
for i in range(points.shape[0]):
- txt = 'p'+str(i+1)
+ txt = 'p' + str(i + 1)
ax1.annotate(txt, (points[i, 0], points[i, 1]))
if scoreExploration is not None:
for i in range(allCandidates.shape[0]):
@@ -1429,14 +1463,14 @@ class Engine():
# -----------------------------------------
# --------- EXPLOITATION METHODS ----------
# -----------------------------------------
- if exploit_method == 'BayesOptDesign' or\
- exploit_method == 'BayesActDesign':
+ if exploit_method.lower() == 'bayesoptdesign' or \
+ exploit_method.lower() == 'bayesactdesign':
# ------- Calculate Exoploration weight -------
# Compute exploration weight based on trade off scheme
explore_w, exploit_w = self.tradeoff_weights(tradeoff_scheme,
- old_EDX,
- old_EDY)
+ old_EDX,
+ old_EDY)
print(f"\n Exploration weight={explore_w:0.3f} "
f"Exploitation weight={exploit_w:0.3f}\n")
@@ -1455,19 +1489,19 @@ class Engine():
# Split the candidates in groups for multiprocessing
split_cand = np.array_split(
candidates, n_cand_groups, axis=0
- )
- # print(candidates)
- # print(split_cand)
+ )
+ # print(candidates)
+ # print(split_cand)
if self.parallel:
results = Parallel(n_jobs=-1, backend='multiprocessing')(
delayed(self.run_util_func)(
exploit_method, split_cand[i], i, sigma2, var, X_MC)
- for i in range(n_cand_groups))
+ for i in range(n_cand_groups))
else:
results = []
for i in range(n_cand_groups):
results.append(self.run_util_func(exploit_method, split_cand[i], i, sigma2, var, X_MC))
-
+
# Retrieve the results and append them
U_J_d = np.concatenate([results[NofE][1] for NofE in
range(n_cand_groups)])
@@ -1489,29 +1523,28 @@ class Engine():
# ------- Calculate Total score -------
# ------- Trade off between EXPLORATION & EXPLOITATION -------
# Accumulate the samples
- finalCandidates = np.concatenate((allCandidates, candidates), axis = 0)
- finalCandidates = np.unique(finalCandidates, axis = 0)
-
+ finalCandidates = np.concatenate((allCandidates, candidates), axis=0)
+ finalCandidates = np.unique(finalCandidates, axis=0)
+
# Calculations take into account both exploration and exploitation
# samples without duplicates
totalScore = np.zeros(finalCandidates.shape[0])
- #self.totalScore = totalScore
-
+ # self.totalScore = totalScore
+
for cand_idx in range(finalCandidates.shape[0]):
# find candidate indices
idx1 = np.where(allCandidates == finalCandidates[cand_idx])[0]
idx2 = np.where(candidates == finalCandidates[cand_idx])[0]
-
+
# exploration
- if idx1 != []:
+ if idx1.shape[0] > 0:
idx1 = idx1[0]
totalScore[cand_idx] += explore_w * scoreExploration[idx1]
-
+
# exploitation
- if idx2 != []:
+ if idx2.shape[0] > 0:
idx2 = idx2[0]
totalScore[cand_idx] += exploit_w * norm_U_J_d[idx2]
-
# Total score
totalScore = exploit_w * norm_U_J_d
@@ -1547,23 +1580,23 @@ class Engine():
Xnew = finalCandidates[sorted_idxtotalScore[:n_new_samples]]
- elif exploit_method == 'VarOptDesign':
+ elif exploit_method.lower() == 'varoptdesign':
# ------- EXPLOITATION: VarOptDesign -------
UtilMethod = var
# ------- Calculate Exoploration weight -------
# Compute exploration weight based on trade off scheme
explore_w, exploit_w = self.tradeoff_weights(tradeoff_scheme,
- old_EDX,
- old_EDY)
+ old_EDX,
+ old_EDY)
print(f"\nweightExploration={explore_w:0.3f} "
f"weightExploitation={exploit_w:0.3f}")
# Generate candidate samples from Exploration class
nMeasurement = old_EDY[OutputNames[0]].shape[1]
-
- # print(UtilMethod)
-
+
+ # print(UtilMethod)
+
# Find sensitive region
if UtilMethod == 'LOOCV':
LCerror = self.MetaModel.LCerror
@@ -1575,12 +1608,12 @@ class Engine():
LCerror[y_key][key])
ExploitScore = np.max(np.max(allModifiedLOO, axis=1), axis=1)
- # print(allModifiedLOO.shape)
+ # print(allModifiedLOO.shape)
elif UtilMethod in ['EIGF', 'ALM']:
# ----- All other in ['EIGF', 'ALM'] -----
# Initilize the ExploitScore array
- ExploitScore = np.zeros((len(old_EDX), len(OutputNames)))
+ # ExploitScore = np.zeros((len(old_EDX), len(OutputNames)))
# Split the candidates in groups for multiprocessing
if explore_method != 'Voronoi':
@@ -1630,9 +1663,9 @@ class Engine():
# Normalize U_J_d
ExploitScore = ExploitScore / np.sum(ExploitScore)
totalScore = exploit_w * ExploitScore
- # print(totalScore.shape)
- # print(explore_w)
- # print(scoreExploration.shape)
+ # print(totalScore.shape)
+ # print(explore_w)
+ # print(scoreExploration.shape)
totalScore += explore_w * scoreExploration
temp = totalScore.copy()
@@ -1654,7 +1687,7 @@ class Engine():
# select the requested number of samples
Xnew[i] = newSamples[np.argmax(maxminScore)]
- elif exploit_method == 'alphabetic':
+ elif exploit_method.lower() == 'alphabetic':
# ------- EXPLOITATION: ALPHABETIC -------
Xnew = self.util_AlphOptDesign(allCandidates, var)
@@ -1676,7 +1709,7 @@ class Engine():
raise NameError('The requested design method is not available.')
print("\n")
- print("\nRun No. {}:".format(old_EDX.shape[0]+1))
+ print("\nRun No. {}:".format(old_EDX.shape[0] + 1))
print("Xnew:\n", Xnew)
# TODO: why does it also return None?
@@ -1694,7 +1727,7 @@ class Engine():
Arguments
---------
- NCandidate : int
+ candidates : int?
Number of candidate points to be searched
var : string
@@ -1705,7 +1738,7 @@ class Engine():
X_new : array of shape (1, n_params)
The new sampling location in the input space.
"""
- MetaModelOrig = self # TODO: this doesn't fully seem correct?
+ MetaModelOrig = self # TODO: this doesn't fully seem correct?
n_new_samples = MetaModelOrig.ExpDesign.n_new_samples
NCandidate = candidates.shape[0]
@@ -1713,7 +1746,7 @@ class Engine():
OutputName = self.out_names[0]
# To avoid changes ub original aPCE object
- MetaModel = deepcopy(MetaModelOrig)
+ # MetaModel = deepcopy(MetaModelOrig)
# Old Experimental design
oldExpDesignX = self.ExpDesign.X
@@ -1722,19 +1755,14 @@ class Engine():
# Suggestion: Go for the one with the highest LOO error
# TODO: this is just a patch, need to look at again!
Scores = list(self.MetaModel.score_dict['b_1'][OutputName].values())
- #print(Scores)
- #print(self.MetaModel.score_dict)
- #print(self.MetaModel.score_dict.values())
- #print(self.MetaModel.score_dict['b_1'].values())
- #print(self.MetaModel.score_dict['b_1'][OutputName].values())
- ModifiedLOO = [1-score for score in Scores]
+ ModifiedLOO = [1 - score for score in Scores]
outIdx = np.argmax(ModifiedLOO)
# Initialize Phi to save the criterion's values
- Phi = np.zeros((NCandidate))
+ Phi = np.zeros(NCandidate)
# TODO: also patched here
- BasisIndices = self.MetaModel.basis_dict['b_1'][OutputName]["y_"+str(outIdx+1)]
+ BasisIndices = self.MetaModel.basis_dict['b_1'][OutputName]["y_" + str(outIdx + 1)]
P = len(BasisIndices)
# ------ Old Psi ------------
@@ -1753,10 +1781,9 @@ class Engine():
# Information matrix
PsiTPsi = np.dot(Psi_cand.T, Psi_cand)
- M = PsiTPsi / (len(oldExpDesignX)+1)
+ M = PsiTPsi / (len(oldExpDesignX) + 1)
- if np.linalg.cond(PsiTPsi) > 1e-12 \
- and np.linalg.cond(PsiTPsi) < 1 / sys.float_info.epsilon:
+ if 1e-12 < np.linalg.cond(PsiTPsi) < 1 / sys.float_info.epsilon:
# faster
invM = linalg.solve(M, sparse.eye(PsiTPsi.shape[0]).toarray())
else:
@@ -1768,7 +1795,7 @@ class Engine():
# D-Opt
if var.lower() == 'd-opt':
- Phi[idx] = (np.linalg.det(invM)) ** (1/P)
+ Phi[idx] = (np.linalg.det(invM)) ** (1 / P)
# A-Opt
elif var.lower() == 'a-opt':
@@ -1779,9 +1806,9 @@ class Engine():
Phi[idx] = np.linalg.cond(M)
else:
- # print(var.lower())
+ # print(var.lower())
raise Exception('The optimality criterion you requested has '
- 'not been implemented yet!')
+ 'not been implemented yet!')
# find an optimal point subset to add to the initial design
# by minimization of the Phi
@@ -1794,7 +1821,7 @@ class Engine():
# -------------------------------------------------------------------------
def _normpdf(self, y_hat_pce, std_pce, obs_data, total_sigma2s,
- rmse=None):
+ rmse=None):
"""
Calculated gaussian likelihood for given y+std based on given obs+sigma
# TODO: is this understanding correct?
@@ -1826,7 +1853,7 @@ class Engine():
for idx, out in enumerate(self.out_names):
# (Meta)Model Output
- # print(y_hat_pce[out])
+ # print(y_hat_pce[out])
nsamples, nout = y_hat_pce[out].shape
# Prepare data and remove NaN
@@ -1841,9 +1868,9 @@ class Engine():
# Surrogate error if valid dataset is given.
if rmse is not None:
- tot_sigma2s += rmse[out]**2
+ tot_sigma2s += rmse[out] ** 2
else:
- tot_sigma2s += np.mean(std_pce[out])**2
+ tot_sigma2s += np.mean(std_pce[out]) ** 2
likelihoods *= stats.multivariate_normal.pdf(
y_hat_pce[out], data, np.diag(tot_sigma2s),
@@ -1870,8 +1897,8 @@ class Engine():
# TODO: Evaluate MetaModel on the experimental design and ValidSet
OutputRS, stdOutputRS = MetaModel.eval_metamodel(samples=samples)
- logLik_data = np.zeros((n_samples))
- logLik_model = np.zeros((n_samples))
+ logLik_data = np.zeros(n_samples)
+ logLik_model = np.zeros(n_samples)
# Loop over the outputs
for idx, out in enumerate(output_names):
@@ -1892,7 +1919,6 @@ class Engine():
covMatrix_data = np.diag(tot_sigma2s)
for i, sample in enumerate(samples):
-
# Simulation run
y_m = model_outputs[out][i]
@@ -1901,22 +1927,22 @@ class Engine():
# CovMatrix with the surrogate error
# covMatrix = np.diag(stdOutputRS[out][i]**2)
- covMatrix = np.diag((y_m-y_m_hat)**2)
+ # covMatrix = np.diag((y_m - y_m_hat) ** 2)
covMatrix = np.diag(
- np.mean((model_outputs[out]-OutputRS[out]), axis=0)**2
- )
+ np.mean((model_outputs[out] - OutputRS[out]), axis=0) ** 2
+ )
# Compute likelilhood output vs data
logLik_data[i] += logpdf(
y_m_hat, data, covMatrix_data
- )
+ )
# Compute likelilhood output vs surrogate
logLik_model[i] += logpdf(y_m_hat, y_m, covMatrix)
# Weight
logLik_data -= logBME
- weights = np.exp(logLik_model+logLik_data)
+ weights = np.exp(logLik_model + logLik_data)
return np.log(np.mean(weights))
@@ -1942,7 +1968,7 @@ class Engine():
"""
# Initialization
- newpath = (r'Outputs_SeqPosteriorComparison/posterior')
+ newpath = r'Outputs_SeqPosteriorComparison/posterior'
os.makedirs(newpath, exist_ok=True)
bound_tuples = self.ExpDesign.bound_tuples
@@ -1984,7 +2010,6 @@ class Engine():
return figPosterior
-
# -------------------------------------------------------------------------
def _BME_Calculator(self, obs_data, sigma2Dict, rmse=None):
"""
@@ -2006,7 +2031,8 @@ class Engine():
"""
# Initializations
- if hasattr(self, 'valid_likelihoods'):
+ # TODO: this just does not make sense, recheck from old commits
+ if self.valid_likelihoods is not None:
valid_likelihoods = self.valid_likelihoods
else:
valid_likelihoods = []
@@ -2014,7 +2040,7 @@ class Engine():
post_snapshot = self.ExpDesign.post_snapshot
if post_snapshot or valid_likelihoods.shape[0] != 0:
- newpath = (r'Outputs_SeqPosteriorComparison/likelihood_vs_ref')
+ newpath = r'Outputs_SeqPosteriorComparison/likelihood_vs_ref'
os.makedirs(newpath, exist_ok=True)
SamplingMethod = 'random'
@@ -2027,7 +2053,7 @@ class Engine():
# Generate samples for Monte Carlo simulation
X_MC = self.ExpDesign.generate_samples(
MCsize, SamplingMethod
- )
+ )
# Monte Carlo simulation for the candidate design
Y_MC, std_MC = self.MetaModel.eval_metamodel(samples=X_MC)
@@ -2036,10 +2062,10 @@ class Engine():
# simulation results via PCE with candidate design)
Likelihoods = self._normpdf(
Y_MC, std_MC, obs_data, sigma2Dict, rmse
- )
+ )
# Check the Effective Sample Size (1000<ESS<MCsize)
- ESS = 1 / np.sum(np.square(Likelihoods/np.sum(Likelihoods)))
+ ESS = 1 / np.sum(np.square(Likelihoods / np.sum(Likelihoods)))
# Enlarge sample size if it doesn't fulfill the criteria
if (ESS > MCsize) or (ESS < 1):
@@ -2052,7 +2078,7 @@ class Engine():
unif = np.random.rand(1, MCsize)[0]
# Reject the poorly performed prior
- accepted = (Likelihoods/np.max(Likelihoods)) >= unif
+ accepted = (Likelihoods / np.max(Likelihoods)) >= unif
X_Posterior = X_MC[accepted]
# ------------------------------------------------------------
@@ -2068,16 +2094,17 @@ class Engine():
postExpLikelihoods = np.mean(np.log(Likelihoods[accepted]))
# Posterior-based expectation of prior densities
- postExpPrior = np.mean(
- np.log(self.ExpDesign.JDist.pdf(X_Posterior.T))
- )
+ # TODO: this is commented out, as it is not used again
+ # postExpPrior = np.mean(
+ # np.log(self.ExpDesign.JDist.pdf(X_Posterior.T))
+ # )
# Calculate Kullback-Leibler Divergence
# KLD = np.mean(np.log(Likelihoods[Likelihoods!=0])- logBME)
KLD = postExpLikelihoods - logBME
# Information Entropy based on Entropy paper Eq. 38
- infEntropy = logBME - postExpPrior - postExpLikelihoods
+ # infEntropy = logBME - postExpPrior - postExpLikelihoods
# If post_snapshot is True, plot likelihood vs refrence
if post_snapshot or valid_likelihoods:
@@ -2086,10 +2113,10 @@ class Engine():
ref_like = np.log(valid_likelihoods[(valid_likelihoods > 0)])
est_like = np.log(Likelihoods[Likelihoods > 0])
distHellinger = hellinger_distance(ref_like, est_like)
-
+
idx = len([name for name in os.listdir(newpath) if 'Likelihoods_'
in name and os.path.isfile(os.path.join(newpath, name))])
-
+
fig, ax = plt.subplots()
try:
sns.kdeplot(np.log(valid_likelihoods[valid_likelihoods > 0]),
@@ -2130,7 +2157,7 @@ class Engine():
'n_walkers': 30,
'moves': emcee.moves.KDEMove(),
'verbose': False
- }
+ }
# ----- Define the discrepancy model -------
# TODO: check with Farid if this first line is how it should be
@@ -2142,13 +2169,13 @@ class Engine():
# # -- (Option B) --
DiscrepancyOpts = Discrepancy('')
DiscrepancyOpts.type = 'Gaussian'
- DiscrepancyOpts.parameters = obs_data**2
+ DiscrepancyOpts.parameters = obs_data ** 2
BayesOpts.Discrepancy = DiscrepancyOpts
# Start the calibration/inference
Bayes_PCE = BayesOpts.create_inference()
X_Posterior = Bayes_PCE.posterior_df.values
- return (logBME, KLD, X_Posterior, Likelihoods, distHellinger)
+ return logBME, KLD, X_Posterior, Likelihoods, distHellinger
# -------------------------------------------------------------------------
def _validError(self):
@@ -2180,14 +2207,14 @@ class Engine():
sample_weight=None,
squared=False)
# Validation error
- valid_error[key] = (rms_error[key]**2)
+ valid_error[key] = (rms_error[key] ** 2)
valid_error[key] /= np.var(valid_model_runs[key], ddof=1, axis=0)
# Print a report table
print("\n>>>>> Updated Errors of {} <<<<<".format(key))
print("\nIndex | RMSE | Validation Error")
- print('-'*35)
- print('\n'.join(f'{i+1} | {k:.3e} | {j:.3e}' for i, (k, j)
+ print('-' * 35)
+ print('\n'.join(f'{i + 1} | {k:.3e} | {j:.3e}' for i, (k, j)
in enumerate(zip(rms_error[key],
valid_error[key]))))
@@ -2213,13 +2240,12 @@ class Engine():
# Compute the root mean squared error
for output in self.out_names:
-
# Compute the error between mean and std of MetaModel and OrigModel
RMSE_Mean = mean_squared_error(
self.Model.mc_reference['mean'], pce_means[output], squared=False
- )
+ )
RMSE_std = mean_squared_error(
self.Model.mc_reference['std'], pce_stds[output], squared=False
- )
+ )
return RMSE_Mean, RMSE_std
diff --git a/src/bayesvalidrox/surrogate_models/exp_designs.py b/src/bayesvalidrox/surrogate_models/exp_designs.py
index fa03fe17d96fb2c1f19546b7b72fb2fd6dd1c13a..ce1745903c776049c797e5585b60d45463d93325 100644
--- a/src/bayesvalidrox/surrogate_models/exp_designs.py
+++ b/src/bayesvalidrox/surrogate_models/exp_designs.py
@@ -4,18 +4,17 @@
Experimental design with associated sampling methods
"""
-import numpy as np
-import math
import itertools
+import math
+
import chaospy
-import scipy.stats as st
-from tqdm import tqdm
import h5py
-import os
+import numpy as np
from .apoly_construction import apoly_construction
from .input_space import InputSpace
+
# -------------------------------------------------------------------------
def check_ranges(theta, ranges):
"""
@@ -26,12 +25,12 @@ def check_ranges(theta, ranges):
theta : array
Proposed parameter set.
ranges : nested list
- List of the praremeter ranges.
+ The parameter ranges.
Returns
-------
c : bool
- If it lies in the given range, it return True else False.
+ If it lies in the given range, it returns True else False.
"""
c = True
@@ -52,7 +51,7 @@ class ExpDesigns(InputSpace):
Attributes
----------
- Input : obj
+ input_object : obj
Input object containing the parameter marginals, i.e. name,
distribution type and distribution parameters or available raw data.
meta_Model_type : str
@@ -143,15 +142,17 @@ class ExpDesigns(InputSpace):
- K-Opt (K-Optimality)
"""
- def __init__(self, Input, meta_Model_type='pce',
- sampling_method='random', hdf5_file=None,
- n_new_samples=1, n_max_samples=None, mod_LOO_threshold=1e-16,
- tradeoff_scheme=None, n_canddidate=1, explore_method='random',
- exploit_method='Space-filling', util_func='Space-filling',
- n_cand_groups=4, n_replication=1, post_snapshot=False,
- step_snapshot=1, max_a_post=[], adapt_verbose=False, max_func_itr=1):
+ def __init__(self, input_object, meta_Model_type='pce', sampling_method='random', hdf5_file=None,
+ n_new_samples=1, n_max_samples=None, mod_LOO_threshold=1e-16, tradeoff_scheme=None, n_canddidate=1,
+ explore_method='random', exploit_method='Space-filling', util_func='Space-filling', n_cand_groups=4,
+ n_replication=1, post_snapshot=False, step_snapshot=1, max_a_post=None, adapt_verbose=False,
+ max_func_itr=1):
- self.InputObj = Input
+ super().__init__(input_object, meta_Model_type)
+ if max_a_post is None:
+ max_a_post = []
+
+ self.InputObj = input_object
self.meta_Model_type = meta_Model_type
self.sampling_method = sampling_method
self.hdf5_file = hdf5_file
@@ -170,17 +171,20 @@ class ExpDesigns(InputSpace):
self.max_a_post = max_a_post
self.adapt_verbose = adapt_verbose
self.max_func_itr = max_func_itr
-
+
# Other
self.apce = None
+ self.n_init_samples = None
+ self.n_samples = None
self.ndim = None
-
+ self.X = None
+ self.Y = None
+
# Init
self.check_valid_inputs()
-
+
# -------------------------------------------------------------------------
- def generate_samples(self, n_samples, sampling_method='random',
- transform=False):
+ def generate_samples(self, n_samples, sampling_method='random'):
"""
Generates samples with given sampling method
@@ -190,9 +194,6 @@ class ExpDesigns(InputSpace):
Number of requested samples.
sampling_method : str, optional
Sampling method. The default is `'random'`.
- transform : bool, optional
- Transformation via an isoprobabilistic transformation method. The
- default is `False`.
Returns
-------
@@ -203,17 +204,14 @@ class ExpDesigns(InputSpace):
try:
samples = chaospy.generate_samples(
int(n_samples), domain=self.origJDist, rule=sampling_method
- )
+ )
except:
samples = self.random_sampler(int(n_samples)).T
return samples.T
-
-
# -------------------------------------------------------------------------
- def generate_ED(self, n_samples, transform=False,
- max_pce_deg=None):
+ def generate_ED(self, n_samples, max_pce_deg=None):
"""
Generates experimental designs (training set) with the given method.
@@ -221,10 +219,6 @@ class ExpDesigns(InputSpace):
----------
n_samples : int
Number of requested training points.
- sampling_method : str, optional
- Sampling method. The default is `'random'`.
- transform : bool, optional
- Isoprobabilistic transformation. The default is `False`.
max_pce_deg : int, optional
Maximum PCE polynomial degree. The default is `None`.
@@ -233,21 +227,23 @@ class ExpDesigns(InputSpace):
None
"""
- if n_samples <0:
+ if n_samples < 0:
raise ValueError('A negative number of samples cannot be created. Please provide positive n_samples')
n_samples = int(n_samples)
-
- if not hasattr(self, 'n_init_samples'):
+
+ if self.n_init_samples is None:
self.n_init_samples = n_samples
# Generate the samples based on requested method
self.init_param_space(max_pce_deg)
+ samples = None
sampling_method = self.sampling_method
# Pass user-defined samples as ED
if sampling_method == 'user':
- if not hasattr(self, 'X'):
- raise AttributeError('User-defined sampling cannot proceed as no samples provided. Please add them to this class as attribute X')
+ if self.X is None:
+ raise AttributeError('User-defined sampling cannot proceed as no samples provided. Please add them to '
+ 'this class as attribute X')
if not self.X.ndim == 2:
raise AttributeError('The provided samples shuld have 2 dimensions')
samples = self.X
@@ -279,7 +275,7 @@ class ExpDesigns(InputSpace):
rule=sampling_method).T
self.X = samples
-
+
def read_from_file(self, out_names):
"""
Reads in the ExpDesign from a provided h5py file and saves the results.
@@ -294,7 +290,7 @@ class ExpDesigns(InputSpace):
None.
"""
- if self.hdf5_file == None:
+ if self.hdf5_file is None:
raise AttributeError('ExpDesign cannot be read in, please provide hdf5 file first')
# Read hdf5 file
@@ -331,11 +327,9 @@ class ExpDesigns(InputSpace):
f.close()
print(f'Experimental Design is read in from file {self.hdf5_file}')
print('')
-
-
# -------------------------------------------------------------------------
- def random_sampler(self, n_samples, max_deg = None):
+ def random_sampler(self, n_samples, max_deg=None):
"""
Samples the given raw data randomly.
@@ -355,10 +349,10 @@ class ExpDesigns(InputSpace):
The sampling locations in the input space.
"""
- if not hasattr(self, 'raw_data'):
+ if self.raw_data is None:
self.init_param_space(max_deg)
else:
- if np.array(self.raw_data).ndim !=2:
+ if np.array(self.raw_data).ndim != 2:
raise AttributeError('The given raw data for sampling should have two dimensions')
samples = np.zeros((n_samples, self.ndim))
sample_size = self.raw_data.shape[1]
@@ -371,10 +365,18 @@ class ExpDesigns(InputSpace):
# store the raw data with given random indices
samples[:, pa_idx] = self.raw_data[pa_idx, rand_idx]
else:
+ if self.JDist is None:
+ raise AttributeError('Sampling cannot proceed, build ExpDesign with max_deg != 0 to create JDist!')
try:
+ # Use resample if JDist is of type gaussian_kde
samples = self.JDist.resample(int(n_samples)).T
except AttributeError:
+ # Use sample if JDist is of type chaospy.J
samples = self.JDist.sample(int(n_samples)).T
+ # If there is only one input transform the samples
+ if self.ndim == 1:
+ samples = np.swapaxes(np.atleast_2d(samples), 0, 1)
+
# Check if all samples are in the bound_tuples
for idx, param_set in enumerate(samples):
if not check_ranges(param_set, self.bound_tuples):
@@ -384,7 +386,7 @@ class ExpDesigns(InputSpace):
except:
proposed_sample = self.JDist.resample(1).T[0]
while not check_ranges(proposed_sample,
- self.bound_tuples):
+ self.bound_tuples):
try:
proposed_sample = chaospy.generate_samples(
1, domain=self.JDist, rule='random').T[0]
@@ -414,49 +416,67 @@ class ExpDesigns(InputSpace):
Collocation points.
"""
-
- if not hasattr(self, 'raw_data'):
+
+ if self.raw_data is None:
self.init_param_space(max_deg)
raw_data = self.raw_data
# Guess the closest degree to self.n_samples
def M_uptoMax(deg):
+ """
+ ??
+ Parameters
+ ----------
+ deg : int
+ Degree.
+
+ Returns
+ -------
+ list of ..?
+ """
result = []
- for d in range(1, deg+1):
- result.append(math.factorial(self.ndim+d) //
+ for d in range(1, deg + 1):
+ result.append(math.factorial(self.ndim + d) //
(math.factorial(self.ndim) * math.factorial(d)))
return np.array(result)
- #print(M_uptoMax(max_deg))
- #print(np.where(M_uptoMax(max_deg) > n_samples)[0])
guess_Deg = np.where(M_uptoMax(max_deg) > n_samples)[0][0]
- c_points = np.zeros((guess_Deg+1, self.ndim))
+ c_points = np.zeros((guess_Deg + 1, self.ndim))
def PolynomialPa(parIdx):
+ """
+ ???
+ Parameters
+ ----------
+ parIdx
+
+ Returns
+ -------
+
+ """
return apoly_construction(self.raw_data[parIdx], max_deg)
for i in range(self.ndim):
- poly_coeffs = PolynomialPa(i)[guess_Deg+1][::-1]
+ poly_coeffs = PolynomialPa(i)[guess_Deg + 1][::-1]
c_points[:, i] = np.trim_zeros(np.roots(poly_coeffs))
# Construction of optimal integration points
- Prod = itertools.product(np.arange(1, guess_Deg+2), repeat=self.ndim)
+ Prod = itertools.product(np.arange(1, guess_Deg + 2), repeat=self.ndim)
sort_dig_unique_combos = np.array(list(filter(lambda x: x, Prod)))
# Ranking relatively mean
- Temp = np.empty(shape=[0, guess_Deg+1])
+ Temp = np.empty(shape=[0, guess_Deg + 1])
for j in range(self.ndim):
- s = abs(c_points[:, j]-np.mean(raw_data[j]))
+ s = abs(c_points[:, j] - np.mean(raw_data[j]))
Temp = np.append(Temp, [s], axis=0)
temp = Temp.T
index_CP = np.sort(temp, axis=0)
- sort_cpoints = np.empty((0, guess_Deg+1))
+ sort_cpoints = np.empty((0, guess_Deg + 1))
for j in range(self.ndim):
- #print(index_CP[:, j])
sort_cp = c_points[index_CP[:, j], j]
sort_cpoints = np.vstack((sort_cpoints, sort_cp))
@@ -464,8 +484,9 @@ class ExpDesigns(InputSpace):
sort_unique_combos = np.empty(shape=[0, self.ndim])
for i in range(len(sort_dig_unique_combos)):
sort_un_comb = []
+ sort_uni_comb = None
for j in range(self.ndim):
- SortUC = sort_cpoints[j, sort_dig_unique_combos[i, j]-1]
+ SortUC = sort_cpoints[j, sort_dig_unique_combos[i, j] - 1]
sort_un_comb.append(SortUC)
sort_uni_comb = np.asarray(sort_un_comb)
sort_unique_combos = np.vstack((sort_unique_combos, sort_uni_comb))
diff --git a/src/bayesvalidrox/surrogate_models/exploration.py b/src/bayesvalidrox/surrogate_models/exploration.py
index 6abb652f145fadb410ecf8f987142e8ceb544a41..67decae2bfef6d397bafbc10abdcb35ff77a63a3 100644
--- a/src/bayesvalidrox/surrogate_models/exploration.py
+++ b/src/bayesvalidrox/surrogate_models/exploration.py
@@ -33,6 +33,7 @@ class Exploration:
def __init__(self, ExpDesign, n_candidate,
mc_criterion='mc-intersite-proj-th'):
+ self.closestPoints = None
self.ExpDesign = ExpDesign
self.n_candidate = n_candidate
self.mc_criterion = mc_criterion
diff --git a/src/bayesvalidrox/surrogate_models/input_space.py b/src/bayesvalidrox/surrogate_models/input_space.py
index 4e010d66f2933ec243bad756d8f2c5454808d802..c534d34b22cbdc9421b5b59c91054032577df4c6 100644
--- a/src/bayesvalidrox/surrogate_models/input_space.py
+++ b/src/bayesvalidrox/surrogate_models/input_space.py
@@ -9,6 +9,7 @@ import chaospy
import scipy.stats as st
+# noinspection SpellCheckingInspection
class InputSpace:
"""
This class generates the input space for the metamodel from the
@@ -24,24 +25,32 @@ class InputSpace:
"""
- def __init__(self, Input, meta_Model_type='pce'):
- self.InputObj = Input
+ def __init__(self, input_object, meta_Model_type='pce'):
+ self.InputObj = input_object
self.meta_Model_type = meta_Model_type
-
+
# Other
self.apce = None
+ self.bound_tuples = None
+ self.input_data_given = None
+ self.JDist = None
+ self.MCSize = None
self.ndim = None
-
+ self.origJDist = None
+ self.par_names = None
+ self.poly_types = None
+ self.prior_space = None
+ self.raw_data = None
+
# Init
self.check_valid_inputs()
-
-
- def check_valid_inputs(self)-> None:
+
+ def check_valid_inputs(self) -> None:
"""
Check if the given InputObj is valid to use for further calculations:
- Has some Marginals
- Marginals have valid priors
- All Marginals given as the same type (samples vs dist)
+ 1) Has some Marginals
+ 2) The Marginals have valid priors
+ 3) All Marginals given as the same type (samples vs dist)
Returns
-------
@@ -50,7 +59,7 @@ class InputSpace:
"""
Inputs = self.InputObj
self.ndim = len(Inputs.Marginals)
-
+
# Check if PCE or aPCE metamodel is selected.
# TODO: test also for 'pce'??
if self.meta_Model_type.lower() == 'apce':
@@ -59,26 +68,22 @@ class InputSpace:
self.apce = False
# check if marginals given
- if not self.ndim >=1:
+ if not self.ndim >= 1:
raise AssertionError('Cannot build distributions if no marginals are given')
-
+
# check that each marginal is valid
for marginals in Inputs.Marginals:
if len(marginals.input_data) == 0:
- if marginals.dist_type == None:
+ if marginals.dist_type is None:
raise AssertionError('Not all marginals were provided priors')
- break
- if np.array(marginals.input_data).shape[0] and (marginals.dist_type != None):
+ if np.array(marginals.input_data).shape[0] and (marginals.dist_type is not None):
raise AssertionError('Both samples and distribution type are given. Please choose only one.')
- break
-
+
# Check if input is given as dist or input_data.
self.input_data_given = -1
for marg in Inputs.Marginals:
- #print(self.input_data_given)
size = np.array(marg.input_data).shape[0]
- #print(f'Size: {size}')
- if size and abs(self.input_data_given) !=1:
+ if size and abs(self.input_data_given) != 1:
self.input_data_given = 2
break
if (not size) and self.input_data_given > 0:
@@ -88,11 +93,10 @@ class InputSpace:
self.input_data_given = 0
if size:
self.input_data_given = 1
-
+
if self.input_data_given == 2:
raise AssertionError('Distributions cannot be built as the priors have different types')
-
-
+
# Get the bounds if input_data are directly defined by user:
if self.input_data_given:
for i in range(self.ndim):
@@ -100,8 +104,6 @@ class InputSpace:
up_bound = np.max(Inputs.Marginals[i].input_data)
Inputs.Marginals[i].parameters = [low_bound, up_bound]
-
-
# -------------------------------------------------------------------------
def init_param_space(self, max_deg=None):
"""
@@ -112,7 +114,7 @@ class InputSpace:
max_deg : int, optional
Maximum degree. The default is `None`.
- Creates
+ Returns
-------
raw_data : array of shape (n_params, n_samples)
Raw data.
@@ -122,7 +124,7 @@ class InputSpace:
"""
# Recheck all before running!
self.check_valid_inputs()
-
+
Inputs = self.InputObj
ndim = self.ndim
rosenblatt_flag = Inputs.Rosenblatt
@@ -192,16 +194,17 @@ class InputSpace:
orig_space_dist : object
A chaospy JDist object or a gaussian_kde object.
poly_types : list
- List of polynomial types for the parameters.
+ A list of polynomial types for the parameters.
"""
Inputs = self.InputObj
-
+
all_data = []
all_dist_types = []
orig_joints = []
poly_types = []
-
+ params = None
+
for parIdx in range(self.ndim):
if Inputs.Marginals[parIdx].dist_type is None:
@@ -222,27 +225,27 @@ class InputSpace:
elif 'unif' in dist_type.lower():
polytype = 'legendre'
- if not np.array(params).shape[0]>=2:
+ if not np.array(params).shape[0] >= 2:
raise AssertionError('Distribution has too few parameters!')
dist = chaospy.Uniform(lower=params[0], upper=params[1])
elif 'norm' in dist_type.lower() and \
- 'log' not in dist_type.lower():
- if not np.array(params).shape[0]>=2:
+ 'log' not in dist_type.lower():
+ if not np.array(params).shape[0] >= 2:
raise AssertionError('Distribution has too few parameters!')
polytype = 'hermite'
dist = chaospy.Normal(mu=params[0], sigma=params[1])
elif 'gamma' in dist_type.lower():
polytype = 'laguerre'
- if not np.array(params).shape[0]>=3:
+ if not np.array(params).shape[0] >= 3:
raise AssertionError('Distribution has too few parameters!')
dist = chaospy.Gamma(shape=params[0],
scale=params[1],
shift=params[2])
elif 'beta' in dist_type.lower():
- if not np.array(params).shape[0]>=4:
+ if not np.array(params).shape[0] >= 4:
raise AssertionError('Distribution has too few parameters!')
polytype = 'jacobi'
dist = chaospy.Beta(alpha=params[0], beta=params[1],
@@ -250,29 +253,29 @@ class InputSpace:
elif 'lognorm' in dist_type.lower():
polytype = 'hermite'
- if not np.array(params).shape[0]>=2:
+ if not np.array(params).shape[0] >= 2:
raise AssertionError('Distribution has too few parameters!')
- mu = np.log(params[0]**2/np.sqrt(params[0]**2 + params[1]**2))
- sigma = np.sqrt(np.log(1 + params[1]**2 / params[0]**2))
+ mu = np.log(params[0] ** 2 / np.sqrt(params[0] ** 2 + params[1] ** 2))
+ sigma = np.sqrt(np.log(1 + params[1] ** 2 / params[0] ** 2))
dist = chaospy.LogNormal(mu, sigma)
# dist = chaospy.LogNormal(mu=params[0], sigma=params[1])
elif 'expon' in dist_type.lower():
polytype = 'exponential'
- if not np.array(params).shape[0]>=2:
+ if not np.array(params).shape[0] >= 2:
raise AssertionError('Distribution has too few parameters!')
dist = chaospy.Exponential(scale=params[0], shift=params[1])
elif 'weibull' in dist_type.lower():
polytype = 'weibull'
- if not np.array(params).shape[0]>=3:
+ if not np.array(params).shape[0] >= 3:
raise AssertionError('Distribution has too few parameters!')
dist = chaospy.Weibull(shape=params[0], scale=params[1],
shift=params[2])
else:
message = (f"DistType {dist_type} for parameter"
- f"{parIdx+1} is not available.")
+ f"{parIdx + 1} is not available.")
raise ValueError(message)
if self.input_data_given or self.apce:
@@ -311,6 +314,8 @@ class InputSpace:
----------
X : array of shape (n_samples,n_params)
Samples to be transformed.
+ params : list
+ Parameters for laguerre/gamma-type distribution.
method : string
If transformation method is 'user' transform X, else just pass X.
@@ -321,17 +326,18 @@ class InputSpace:
"""
# Check for built JDist
- if not hasattr(self, 'JDist'):
+ if self.JDist is None:
raise AttributeError('Call function init_param_space first to create JDist')
-
+
# Check if X is 2d
if X.ndim != 2:
raise AttributeError('X should have two dimensions')
-
+
# Check if size of X matches Marginals
- if X.shape[1]!= self.ndim:
- raise AttributeError('The second dimension of X should be the same size as the number of marginals in the InputObj')
-
+ if X.shape[1] != self.ndim:
+ raise AttributeError(
+ 'The second dimension of X should be the same size as the number of marginals in the InputObj')
+
if self.InputObj.Rosenblatt:
self.origJDist, _ = self.build_polytypes(False)
if method == 'user':
@@ -354,8 +360,8 @@ class InputSpace:
if None in disttypes or self.input_data_given or self.apce:
return X
- cdfx = np.zeros((X.shape))
- tr_X = np.zeros((X.shape))
+ cdfx = np.zeros(X.shape)
+ tr_X = np.zeros(X.shape)
for par_i in range(n_params):
@@ -370,11 +376,12 @@ class InputSpace:
# Extract the parameters of the transformation space based on
# polyType
+ inv_cdf = None
if polytype == 'legendre' or disttype == 'uniform':
# Generate Y_Dists based
params_Y = [-1, 1]
dist_Y = st.uniform(loc=params_Y[0],
- scale=params_Y[1]-params_Y[0])
+ scale=params_Y[1] - params_Y[0])
inv_cdf = np.vectorize(lambda x: dist_Y.ppf(x))
elif polytype == 'hermite' or disttype == 'norm':
@@ -383,9 +390,11 @@ class InputSpace:
inv_cdf = np.vectorize(lambda x: dist_Y.ppf(x))
elif polytype == 'laguerre' or disttype == 'gamma':
- if params == None:
+ if params is None:
raise AttributeError('Additional parameters have to be set for the gamma distribution!')
params_Y = [1, params[1]]
+
+ # TOOD: update the call to the gamma function, seems like source code has been changed!
dist_Y = st.gamma(loc=params_Y[0], scale=params_Y[1])
inv_cdf = np.vectorize(lambda x: dist_Y.ppf(x))
diff --git a/src/bayesvalidrox/surrogate_models/inputs.py b/src/bayesvalidrox/surrogate_models/inputs.py
index 094e1066fe008e37288e44750524c5a1370bd7a2..40ae36337fa84c0bb9758e488f983ba29cdb9e77 100644
--- a/src/bayesvalidrox/surrogate_models/inputs.py
+++ b/src/bayesvalidrox/surrogate_models/inputs.py
@@ -4,6 +4,7 @@
Inputs and related marginal distributions
"""
+
class Input:
"""
A class to define the uncertain input parameters.
@@ -20,17 +21,18 @@ class Input:
-------
Marginals can be defined as following:
- >>> Inputs.add_marginals()
- >>> Inputs.Marginals[0].name = 'X_1'
- >>> Inputs.Marginals[0].dist_type = 'uniform'
- >>> Inputs.Marginals[0].parameters = [-5, 5]
+ >>> inputs = Inputs()
+ >>> inputs.add_marginals()
+ >>> inputs.Marginals[0].name = 'X_1'
+ >>> inputs.Marginals[0].dist_type = 'uniform'
+ >>> inputs.Marginals[0].parameters = [-5, 5]
If there is no common data is avaliable, the input data can be given
as following:
- >>> Inputs.add_marginals()
- >>> Inputs.Marginals[0].name = 'X_1'
- >>> Inputs.Marginals[0].input_data = input_data
+ >>> inputs.add_marginals()
+ >>> inputs.Marginals[0].name = 'X_1'
+ >>> inputs.Marginals[0].input_data = [0,0,1,0]
"""
poly_coeffs_flag = True
@@ -63,12 +65,12 @@ class Marginal:
dist_type : string
Name of the distribution. The default is `None`.
parameters : list
- List of the parameters corresponding to the distribution type. The
+ Parameters corresponding to the distribution type. The
default is `None`.
input_data : array
Available input data. The default is `[]`.
moments : list
- List of the moments.
+ Moments of the distribution. The default is `None`.
"""
def __init__(self):
diff --git a/src/bayesvalidrox/surrogate_models/reg_fast_ard.py b/src/bayesvalidrox/surrogate_models/reg_fast_ard.py
index e6883a3edd6d247c219b8be328f5206b75780fbb..fdd0ee7470dc5d0bdccd354cee16f6257ec01b02 100755
--- a/src/bayesvalidrox/surrogate_models/reg_fast_ard.py
+++ b/src/bayesvalidrox/surrogate_models/reg_fast_ard.py
@@ -236,7 +236,7 @@ class RegressionFastARD(LinearModel, RegressorMixin):
self.var_y = False
A = np.PINF * np.ones(n_features)
- active = np.zeros(n_features, dtype=np.bool)
+ active = np.zeros(n_features, dtype=bool)
if self.start is not None and not hasattr(self, 'active_'):
start = self.start
diff --git a/src/bayesvalidrox/surrogate_models/surrogate_models.py b/src/bayesvalidrox/surrogate_models/surrogate_models.py
index ca902f26bef0c45e8befb72ff67313ef09a77603..d8a589dde7973e6c7c86ba6d246e297bf001397a 100644
--- a/src/bayesvalidrox/surrogate_models/surrogate_models.py
+++ b/src/bayesvalidrox/surrogate_models/surrogate_models.py
@@ -4,38 +4,284 @@
Implementation of metamodel as either PC, aPC or GPE
"""
+import copy
+import os
import warnings
-import numpy as np
-import math
-import h5py
+
import matplotlib.pyplot as plt
-from sklearn.preprocessing import MinMaxScaler
+import numpy as np
import scipy as sp
-from scipy.optimize import minimize, NonlinearConstraint, LinearConstraint
-from tqdm import tqdm
-from sklearn.decomposition import PCA as sklearnPCA
-import sklearn.linear_model as lm
-from sklearn.gaussian_process import GaussianProcessRegressor
import sklearn.gaussian_process.kernels as kernels
-import os
+import sklearn.linear_model as lm
from joblib import Parallel, delayed
-import copy
+from scipy.optimize import minimize, NonlinearConstraint
+from sklearn.decomposition import PCA as sklearnPCA
+from sklearn.gaussian_process import GaussianProcessRegressor
+from sklearn.preprocessing import MinMaxScaler
+from tqdm import tqdm
-from .input_space import InputSpace
-from .glexindex import glexindex
+from .apoly_construction import apoly_construction
+from .bayes_linear import VBLinearRegression, EBLinearRegression
from .eval_rec_rule import eval_univ_basis
+from .glexindex import glexindex
+from .input_space import InputSpace
+from .orthogonal_matching_pursuit import OrthogonalMatchingPursuit
from .reg_fast_ard import RegressionFastARD
from .reg_fast_laplace import RegressionFastLaplace
-from .orthogonal_matching_pursuit import OrthogonalMatchingPursuit
-from .bayes_linear import VBLinearRegression, EBLinearRegression
-from .apoly_construction import apoly_construction
+
warnings.filterwarnings("ignore")
# Load the mplstyle
+# noinspection SpellCheckingInspection
plt.style.use(os.path.join(os.path.split(__file__)[0],
'../', 'bayesvalidrox.mplstyle'))
-class MetaModel():
+# noinspection SpellCheckingInspection
+def corr_loocv_error(clf, psi, coeffs, y):
+ """
+ Calculates the corrected LOO error for regression on regressor
+ matrix `psi` that generated the coefficients based on [1] and [2].
+
+ [1] Blatman, G., 2009. Adaptive sparse polynomial chaos expansions for
+ uncertainty propagation and sensitivity analysis (Doctoral
+ dissertation, Clermont-Ferrand 2).
+
+ [2] Blatman, G. and Sudret, B., 2011. Adaptive sparse polynomial chaos
+ expansion based on least angle regression. Journal of computational
+ Physics, 230(6), pp.2345-2367.
+
+ Parameters
+ ----------
+ clf : object
+ Fitted estimator.
+ psi : array of shape (n_samples, n_features)
+ The multivariate orthogonal polynomials (regressor).
+ coeffs : array-like of shape (n_features,)
+ Estimated cofficients.
+ y : array of shape (n_samples,)
+ Target values.
+
+ Returns
+ -------
+ R_2 : float
+ LOOCV Validation score (1-LOOCV erro).
+ residual : array of shape (n_samples,)
+ Residual values (y - predicted targets).
+
+ """
+ psi = np.array(psi, dtype=float)
+
+ # Create PSI_Sparse by removing redundent terms
+ nnz_idx = np.nonzero(coeffs)[0]
+ if len(nnz_idx) == 0:
+ nnz_idx = [0]
+ psi_sparse = psi[:, nnz_idx]
+
+ # NrCoeffs of aPCEs
+ P = len(nnz_idx)
+ # NrEvaluation (Size of experimental design)
+ N = psi.shape[0]
+
+ # Build the projection matrix
+ PsiTPsi = np.dot(psi_sparse.T, psi_sparse)
+
+ if np.linalg.cond(PsiTPsi) > 1e-12: # and \
+ # np.linalg.cond(PsiTPsi) < 1/sys.float_info.epsilon:
+ # faster
+ try:
+ M = sp.linalg.solve(PsiTPsi,
+ sp.sparse.eye(PsiTPsi.shape[0]).toarray())
+ except:
+ raise AttributeError(
+ 'There are too few samples for the corrected loo-cv error. Fit surrogate on at least as many '
+ 'samples as parameters to use this')
+ else:
+ # stabler
+ M = np.linalg.pinv(PsiTPsi)
+
+ # h factor (the full matrix is not calculated explicitly,
+ # only the trace is, to save memory)
+ PsiM = np.dot(psi_sparse, M)
+
+ h = np.sum(np.multiply(PsiM, psi_sparse), axis=1, dtype=np.longdouble) # float128)
+
+ # ------ Calculate Error Loocv for each measurement point ----
+ # Residuals
+ try:
+ residual = clf.predict(psi) - y
+ except:
+ residual = np.dot(psi, coeffs) - y
+
+ # Variance
+ var_y = np.var(y)
+
+ if var_y == 0:
+ # norm_emp_error = 0
+ loo_error = 0
+ LCerror = np.zeros(y.shape)
+ return 1 - loo_error, LCerror
+ else:
+ # norm_emp_error = np.mean(residual ** 2) / var_y
+
+ # LCerror = np.divide(residual, (1-h))
+ LCerror = residual / (1 - h)
+ loo_error = np.mean(np.square(LCerror)) / var_y
+ # if there are NaNs, just return an infinite LOO error (this
+ # happens, e.g., when a strongly underdetermined problem is solved)
+ if np.isnan(loo_error):
+ loo_error = np.inf
+
+ # Corrected Error for over-determined system
+ tr_M = np.trace(M)
+ if tr_M < 0 or abs(tr_M) > 1e6:
+ tr_M = np.trace(np.linalg.pinv(np.dot(psi.T, psi)))
+
+ # Over-determined system of Equation
+ if N > P:
+ T_factor = N / (N - P) * (1 + tr_M)
+
+ # Under-determined system of Equation
+ else:
+ T_factor = np.inf
+
+ corrected_loo_error = loo_error * T_factor
+
+ R_2 = 1 - corrected_loo_error
+
+ return R_2, LCerror
+
+
+def create_psi(basis_indices, univ_p_val):
+ """
+ This function assemble the design matrix Psi from the given basis index
+ set INDICES and the univariate polynomial evaluations univ_p_val.
+
+ Parameters
+ ----------
+ basis_indices : array of shape (n_terms, n_params)
+ Multi-indices of multivariate polynomials.
+ univ_p_val : array of (n_samples, n_params, n_max+1)
+ All univariate regressors up to `n_max`.
+
+ Raises
+ ------
+ ValueError
+ n_terms in arguments do not match.
+
+ Returns
+ -------
+ psi : array of shape (n_samples, n_terms)
+ Multivariate regressors.
+
+ """
+ # Check if BasisIndices is a sparse matrix
+ sparsity = sp.sparse.issparse(basis_indices)
+ if sparsity:
+ basis_indices = basis_indices.toarray()
+
+ # Initialization and consistency checks
+ # number of input variables
+ n_params = univ_p_val.shape[1]
+
+ # Size of the experimental design
+ n_samples = univ_p_val.shape[0]
+
+ # number of basis terms
+ n_terms = basis_indices.shape[0]
+
+ # check that the variables have consistent sizes
+ if n_params != basis_indices.shape[1]:
+ raise ValueError(
+ f"The shapes of basis_indices ({basis_indices.shape[1]}) and "
+ f"univ_p_val ({n_params}) don't match!!"
+ )
+
+ # Preallocate the Psi matrix for performance
+ psi = np.ones((n_samples, n_terms))
+ # Assemble the Psi matrix
+ for m in range(basis_indices.shape[1]):
+ aa = np.where(basis_indices[:, m] > 0)[0]
+ try:
+ basisIdx = basis_indices[aa, m]
+ bb = univ_p_val[:, m, basisIdx].reshape(psi[:, aa].shape)
+ psi[:, aa] = np.multiply(psi[:, aa], bb)
+ except ValueError as err:
+ raise err
+ return psi
+
+
+def gaussian_process_emulator(X, y, nug_term=None, autoSelect=False,
+ varIdx=None):
+ """
+ Fits a Gaussian Process Emulator to the target given the training
+ points.
+
+ Parameters
+ ----------
+ X : array of shape (n_samples, n_params)
+ Training points.
+ y : array of shape (n_samples,)
+ Target values.
+ nug_term : float, optional
+ Nugget term. The default is None, i.e. variance of y.
+ autoSelect : bool, optional
+ Loop over some kernels and select the best. The default is False.
+ varIdx : int, optional
+ The index number. The default is None.
+
+ Returns
+ -------
+ gp : object
+ Fitted estimator.
+
+ """
+
+ nug_term = nug_term if nug_term else np.var(y)
+
+ Kernels = [nug_term * kernels.RBF(length_scale=1.0,
+ length_scale_bounds=(1e-25, 1e15)),
+ nug_term * kernels.RationalQuadratic(length_scale=0.2,
+ alpha=1.0),
+ nug_term * kernels.Matern(length_scale=1.0,
+ length_scale_bounds=(1e-15, 1e5),
+ nu=1.5)]
+
+ # Automatic selection of the kernel
+ if autoSelect:
+ gp = {}
+ BME = []
+ for i, kernel in enumerate(Kernels):
+ gp[i] = GaussianProcessRegressor(kernel=kernel,
+ n_restarts_optimizer=3,
+ normalize_y=False)
+
+ # Fit to data using Maximum Likelihood Estimation
+ gp[i].fit(X, y)
+
+ # Store the MLE as BME score
+ BME.append(gp[i].log_marginal_likelihood())
+
+ gp = gp[np.argmax(BME)]
+
+ else:
+ gp = GaussianProcessRegressor(kernel=Kernels[0],
+ n_restarts_optimizer=3,
+ normalize_y=False)
+ gp.fit(X, y)
+
+ # Compute score
+ if varIdx is not None:
+ Score = gp.score(X, y)
+ print('-' * 50)
+ print(f'Output variable {varIdx}:')
+ print('The estimation of GPE coefficients converged,')
+ print(f'with the R^2 score: {Score:.3f}')
+ print('-' * 50)
+
+ return gp
+
+
+class MetaModel:
"""
Meta (surrogate) model
@@ -82,7 +328,7 @@ class MetaModel():
`'no'`. There are two ways to select number of components: use
percentage of the explainable variance threshold (between 0 and 100)
(Option A) or direct prescription of components' number (Option B):
-
+ >>> MetaModelOpts = MetaModel()
>>> MetaModelOpts.dim_red_method = 'PCA'
>>> MetaModelOpts.var_pca_threshold = 99.999 # Option A
>>> MetaModelOpts.n_pca_components = 12 # Option B
@@ -110,7 +356,7 @@ class MetaModel():
def __init__(self, input_obj, meta_model_type='PCE',
pce_reg_method='OLS', bootstrap_method='fast',
n_bootstrap_itrs=1, pce_deg=1, pce_q_norm=1.0,
- dim_red_method='no', apply_constraints = False,
+ dim_red_method='no', apply_constraints=False,
verbose=False):
self.input_obj = input_obj
@@ -123,8 +369,38 @@ class MetaModel():
self.dim_red_method = dim_red_method
self.apply_constraints = apply_constraints
self.verbose = verbose
-
- def build_metamodel(self, n_init_samples = None) -> None:
+
+ # Other params
+ self.InputSpace = None
+ self.var_pca_threshold = None
+ self.polycoeffs = None
+ self.errorScale = None
+ self.errorclf_poly = None
+ self.errorRegMethod = None
+ self.nlc = None
+ self.univ_p_val = None
+ self.n_pca_components = None
+ self.out_names = None
+ self.allBasisIndices = None
+ self.deg_array = None
+ self.n_samples = None
+ self.CollocationPoints = None
+ self.pca = None
+ self.LCerror = None
+ self.clf_poly = None
+ self.score_dict = None
+ self.basis_dict = None
+ self.coeffs_dict = None
+ self.q_norm_dict = None
+ self.deg_dict = None
+ self.x_scaler = None
+ self.gp_poly = None
+ self.n_params = None
+ self.ndim = None
+ self.init_type = None
+ self.rmse = None
+
+ def build_metamodel(self, n_init_samples=None) -> None:
"""
Builds the parts for the metamodel (polynomes,...) that are neede before fitting.
@@ -134,31 +410,33 @@ class MetaModel():
DESCRIPTION.
"""
-
+
# Generate general warnings
if self.apply_constraints or self.pce_reg_method.lower() == 'ols':
print('There are no estimations of surrogate uncertainty available'
' for the chosen regression options. This might lead to issues'
' in later steps.')
-
+
+ if self.CollocationPoints is None:
+ raise AttributeError('Please provide samples to the metamodel before building it.')
+ self.CollocationPoints = np.array(self.CollocationPoints)
+
# Add InputSpace to MetaModel if it does not have any
- if not hasattr(self, 'InputSpace'):
+ if self.InputSpace is None:
+ if n_init_samples is None:
+ n_init_samples = self.CollocationPoints.shape[0]
self.InputSpace = InputSpace(self.input_obj)
self.InputSpace.n_init_samples = n_init_samples
self.InputSpace.init_param_space(np.max(self.pce_deg))
-
+
self.ndim = self.InputSpace.ndim
-
- if not hasattr(self, 'CollocationPoints'):
- raise AttributeError('Please provide samples to the metamodel before building it.')
-
+
# Transform input samples
# TODO: this is probably not yet correct! Make 'method' variable
- self.CollocationPoints = self.InputSpace.transform(self.CollocationPoints, method='user')
+ self.CollocationPoints = self.InputSpace.transform(self.CollocationPoints, method='user')
-
self.n_params = len(self.input_obj.Marginals)
-
+
# Generate polynomials
if self.meta_model_type.lower() != 'gpe':
self.generate_polynomials(np.max(self.pce_deg))
@@ -183,8 +461,10 @@ class MetaModel():
self.CollocationPoints = np.array(self.CollocationPoints)
self.n_samples, ndim = self.CollocationPoints.shape
if self.ndim != ndim:
- raise AttributeError('The given samples do not match the given number of priors. The samples should be a 2D array of size (#samples, #priors)')
-
+ raise AttributeError(
+ 'The given samples do not match the given number of priors. The samples should be a 2D array of size'
+ '(#samples, #priors)')
+
self.deg_array = self.__select_degree(ndim, self.n_samples)
# Generate all basis indices
@@ -194,15 +474,13 @@ class MetaModel():
if deg not in np.fromiter(keys, dtype=float):
# Generate the polynomial basis indices
for qidx, q in enumerate(self.pce_q_norm):
- basis_indices = glexindex(start=0, stop=deg+1,
+ basis_indices = glexindex(start=0, stop=deg + 1,
dimensions=self.n_params,
cross_truncation=q,
reverse=False, graded=True)
self.allBasisIndices[str(deg)][str(q)] = basis_indices
-
-
- def fit(self, X, y, parallel = True, verbose = False):
+ def fit(self, X: np.array, y: dict, parallel=False, verbose=False):
"""
Fits the surrogate to the given data (samples X, outputs y).
Note here that the samples X should be the transformed samples provided
@@ -214,33 +492,43 @@ class MetaModel():
The parameter value combinations that the model was evaluated at.
y : dict of 2D lists or arrays of shape (#samples, #timesteps)
The respective model evaluations.
+ parallel : bool
+ Set to True to run the training in parallel for various keys.
+ The default is False.
+ verbose : bool
+ Set to True to obtain more information during runtime.
+ The default is False.
Returns
-------
None.
"""
+ # print(X)
+ # print(X.shape)
+ # print(y)
+ # print(y['Z'].shape)
X = np.array(X)
for key in y.keys():
y_val = np.array(y[key])
- if y_val.ndim !=2:
+ if y_val.ndim != 2:
raise ValueError('The given outputs y should be 2D')
y[key] = np.array(y[key])
-
+
# Output names are the same as the keys in y
self.out_names = list(y.keys())
-
+
# Build the MetaModel on the static samples
self.CollocationPoints = X
-
+
# TODO: other option: rebuild every time
- if not hasattr(self, 'deg_array'):
- self.build_metamodel(n_init_samples = X.shape[1])
-
+ if self.deg_array is None:
+ self.build_metamodel(n_init_samples=X.shape[1])
+
# Evaluate the univariate polynomials on InputSpace
if self.meta_model_type.lower() != 'gpe':
- self.univ_p_val = self.univ_basis_vals(self.CollocationPoints)
-
+ self.univ_p_val = self.univ_basis_vals(self.CollocationPoints)
+
# --- Loop through data points and fit the surrogate ---
if verbose:
print(f"\n>>>> Training the {self.meta_model_type} metamodel "
@@ -252,10 +540,10 @@ class MetaModel():
self.n_bootstrap_itrs = 100
# Check if fast version (update coeffs with OLS) is selected.
+ n_comp_dict = {}
+ first_out = {}
if self.bootstrap_method.lower() == 'fast':
fast_bootstrap = True
- first_out = {}
- n_comp_dict = {}
else:
fast_bootstrap = False
@@ -295,8 +583,8 @@ class MetaModel():
# Start transformation
pca, target, n_comp = self.pca_transformation(
Output[b_indices], verbose=False
- )
- self.pca[f'b_{b_i+1}'][key] = pca
+ )
+ self.pca[f'b_{b_i + 1}'][key] = pca
# Store the number of components for fast bootsrtrapping
if fast_bootstrap and b_i == 0:
n_comp_dict[key] = n_comp
@@ -304,39 +592,40 @@ class MetaModel():
target = Output[b_indices]
# Parallel fit regression
+ out = None
if self.meta_model_type.lower() == 'gpe':
# Prepare the input matrix
scaler = MinMaxScaler()
X_S = scaler.fit_transform(X_train_b)
- self.x_scaler[f'b_{b_i+1}'][key] = scaler
+ self.x_scaler[f'b_{b_i + 1}'][key] = scaler
if parallel:
out = Parallel(n_jobs=-1, backend='multiprocessing')(
- delayed(self.gaussian_process_emulator)(
+ delayed(gaussian_process_emulator)(
X_S, target[:, idx]) for idx in
range(target.shape[1]))
else:
- results = map(self.gaussian_process_emulator,
- [X_train_b]*target.shape[1],
+ results = map(gaussian_process_emulator,
+ [X_train_b] * target.shape[1],
[target[:, idx] for idx in
range(target.shape[1])]
)
out = list(results)
for idx in range(target.shape[1]):
- self.gp_poly[f'b_{b_i+1}'][key][f"y_{idx+1}"] = out[idx]
+ self.gp_poly[f'b_{b_i + 1}'][key][f"y_{idx + 1}"] = out[idx]
else:
self.univ_p_val = self.univ_p_val[b_indices]
if parallel and (not fast_bootstrap or b_i == 0):
out = Parallel(n_jobs=-1, backend='multiprocessing')(
- delayed(self.adaptive_regression)(X_train_b,
- target[:, idx],
- idx)
+ delayed(self.adaptive_regression)( # X_train_b,
+ target[:, idx],
+ idx)
for idx in range(target.shape[1]))
elif not parallel and (not fast_bootstrap or b_i == 0):
results = map(self.adaptive_regression,
- [X_train_b]*target.shape[1],
+ # [X_train_b] * target.shape[1],
[target[:, idx] for idx in
range(target.shape[1])],
range(target.shape[1]))
@@ -347,27 +636,26 @@ class MetaModel():
first_out[key] = copy.deepcopy(out)
if b_i > 0 and fast_bootstrap:
-
# fast bootstrap
out = self.update_pce_coeffs(
X_train_b, target, first_out[key])
for i in range(target.shape[1]):
# Create a dict to pass the variables
- self.deg_dict[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['degree']
- self.q_norm_dict[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['qnorm']
- self.coeffs_dict[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['coeffs']
- self.basis_dict[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['multi_indices']
- self.score_dict[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['LOOCVScore']
- self.clf_poly[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['clf_poly']
- #self.LCerror[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['LCerror']
+ self.deg_dict[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['degree']
+ self.q_norm_dict[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['qnorm']
+ self.coeffs_dict[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['coeffs']
+ self.basis_dict[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['multi_indices']
+ self.score_dict[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['LOOCVScore']
+ self.clf_poly[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['clf_poly']
+ # self.LCerror[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['LCerror']
if verbose:
print(f"\n>>>> Training the {self.meta_model_type} metamodel"
" sucessfully completed. <<<<<<\n")
# -------------------------------------------------------------------------
- def update_pce_coeffs(self, X, y, out_dict = None):
+ def update_pce_coeffs(self, X, y, out_dict=None):
"""
Updates the PCE coefficents using only the ordinary least square method
for the fast version of the bootstrapping.
@@ -388,26 +676,26 @@ class MetaModel():
The updated training output dictionary.
"""
+ # TODO: why is X not used here?
# Make a copy
final_out_dict = copy.deepcopy(out_dict)
# Loop over the points
for i in range(y.shape[1]):
-
# Extract nonzero basis indices
nnz_idx = np.nonzero(out_dict[i]['coeffs'])[0]
if len(nnz_idx) != 0:
basis_indices = out_dict[i]['multi_indices']
# Evaluate the multivariate polynomials on CollocationPoints
- psi = self.create_psi(basis_indices, self.univ_p_val)
+ psi = create_psi(basis_indices, self.univ_p_val)
# Calulate the cofficients of surrogate model
updated_out = self.regression(
psi, y[:, i], basis_indices, reg_method='OLS',
sparsity=False
- )
+ )
# Update coeffs in out_dict
final_out_dict[i]['coeffs'][nnz_idx] = updated_out['coeffs']
@@ -425,7 +713,7 @@ class MetaModel():
"""
self.InputSpace = InputSpace(self.input_obj,
- meta_Model_type=self.meta_model_type)
+ meta_Model_type=self.meta_model_type)
# -------------------------------------------------------------------------
def univ_basis_vals(self, samples, n_max=None):
@@ -462,63 +750,6 @@ class MetaModel():
return univ_basis
# -------------------------------------------------------------------------
- def create_psi(self, basis_indices, univ_p_val):
- """
- This function assemble the design matrix Psi from the given basis index
- set INDICES and the univariate polynomial evaluations univ_p_val.
-
- Parameters
- ----------
- basis_indices : array of shape (n_terms, n_params)
- Multi-indices of multivariate polynomials.
- univ_p_val : array of (n_samples, n_params, n_max+1)
- All univariate regressors up to `n_max`.
-
- Raises
- ------
- ValueError
- n_terms in arguments do not match.
-
- Returns
- -------
- psi : array of shape (n_samples, n_terms)
- Multivariate regressors.
-
- """
- # Check if BasisIndices is a sparse matrix
- sparsity = sp.sparse.issparse(basis_indices)
- if sparsity:
- basis_indices = basis_indices.toarray()
-
- # Initialization and consistency checks
- # number of input variables
- n_params = univ_p_val.shape[1]
-
- # Size of the experimental design
- n_samples = univ_p_val.shape[0]
-
- # number of basis terms
- n_terms = basis_indices.shape[0]
-
- # check that the variables have consistent sizes
- if n_params != basis_indices.shape[1]:
- raise ValueError(
- f"The shapes of basis_indices ({basis_indices.shape[1]}) and "
- f"univ_p_val ({n_params}) don't match!!"
- )
-
- # Preallocate the Psi matrix for performance
- psi = np.ones((n_samples, n_terms))
- # Assemble the Psi matrix
- for m in range(basis_indices.shape[1]):
- aa = np.where(basis_indices[:, m] > 0)[0]
- try:
- basisIdx = basis_indices[aa, m]
- bb = univ_p_val[:, m, basisIdx].reshape(psi[:, aa].shape)
- psi[:, aa] = np.multiply(psi[:, aa], bb)
- except ValueError as err:
- raise err
- return psi
# -------------------------------------------------------------------------
def regression(self, X, y, basis_indices, reg_method=None, sparsity=True):
@@ -536,6 +767,8 @@ class MetaModel():
Multi-indices of multivariate polynomials.
reg_method : str, optional
DESCRIPTION. The default is None.
+ sparsity : bool
+ Use with sparsity-inducing training methods. The default is True
Returns
-------
@@ -557,22 +790,23 @@ class MetaModel():
Lambda = 1e-6
# Bayes sparse adaptive aPCE
+ clf_poly = None
if reg_method.lower() == 'ols':
clf_poly = lm.LinearRegression(fit_intercept=False)
elif reg_method.lower() == 'brr':
clf_poly = lm.BayesianRidge(n_iter=1000, tol=1e-7,
fit_intercept=False,
- #normalize=True,
+ # normalize=True,
compute_score=compute_score,
alpha_1=1e-04, alpha_2=1e-04,
lambda_1=Lambda, lambda_2=Lambda)
clf_poly.converged = True
elif reg_method.lower() == 'ard':
- if X.shape[0]<2:
+ if X.shape[0] < 2:
raise ValueError('Regression with ARD can only be performed for more than 2 samples')
clf_poly = lm.ARDRegression(fit_intercept=False,
- #normalize=True,
+ # normalize=True,
compute_score=compute_score,
n_iter=1000, tol=0.0001,
alpha_1=1e-3, alpha_2=1e-3,
@@ -585,14 +819,14 @@ class MetaModel():
n_iter=300, tol=1e-10)
elif reg_method.lower() == 'bcs':
- if X.shape[0]<10:
+ if X.shape[0] < 10:
raise ValueError('Regression with BCS can only be performed for more than 10 samples')
clf_poly = RegressionFastLaplace(fit_intercept=False,
- bias_term=bias_term,
- n_iter=1000, tol=1e-7)
+ bias_term=bias_term,
+ n_iter=1000, tol=1e-7)
elif reg_method.lower() == 'lars':
- if X.shape[0]<10:
+ if X.shape[0] < 10:
raise ValueError('Regression with LARS can only be performed for more than 5 samples')
clf_poly = lm.LassoLarsCV(fit_intercept=False)
@@ -608,29 +842,29 @@ class MetaModel():
elif reg_method.lower() == 'ebl':
clf_poly = EBLinearRegression(optimizer='em')
-
-
+
# Training with constraints automatically uses L2
- if self.apply_constraints:
+ if self.apply_constraints:
# TODO: set the constraints here
# Define the nonlin. constraint
- nlc = NonlinearConstraint(lambda x: np.matmul(X,x),-1,1.1)
+ nlc = NonlinearConstraint(lambda x: np.matmul(X, x), -1, 1.1)
self.nlc = nlc
-
- fun = lambda x: (np.linalg.norm(np.matmul(X, x)-y, ord = 2))**2
- if self.init_type =='zeros':
- res = minimize(fun, np.zeros(X.shape[1]), method = 'trust-constr', constraints = self.nlc)
+
+ fun = lambda x: (np.linalg.norm(np.matmul(X, x) - y, ord=2)) ** 2
+ res = None
+ if self.init_type == 'zeros':
+ res = minimize(fun, np.zeros(X.shape[1]), method='trust-constr', constraints=self.nlc)
if self.init_type == 'nonpi':
clf_poly.fit(X, y)
coeff = clf_poly.coef_
- res = minimize(fun, coeff, method = 'trust-constr', constraints = self.nlc)
-
+ res = minimize(fun, coeff, method='trust-constr', constraints=self.nlc)
+
coeff = np.array(res.x)
clf_poly.coef_ = coeff
clf_poly.X = X
clf_poly.y = y
clf_poly.intercept_ = 0
-
+
# Training without constraints uses chosen regression method
else:
clf_poly.fit(X, y)
@@ -658,76 +892,15 @@ class MetaModel():
return_out_dict['sparePsi'] = sparse_X
return_out_dict['coeffs'] = coeffs
return return_out_dict
-
- # -------------------------------------------------------------------------
- def create_psi(self, basis_indices, univ_p_val):
- """
- This function assemble the design matrix Psi from the given basis index
- set INDICES and the univariate polynomial evaluations univ_p_val.
-
- Parameters
- ----------
- basis_indices : array of shape (n_terms, n_params)
- Multi-indices of multivariate polynomials.
- univ_p_val : array of (n_samples, n_params, n_max+1)
- All univariate regressors up to `n_max`.
-
- Raises
- ------
- ValueError
- n_terms in arguments do not match.
-
- Returns
- -------
- psi : array of shape (n_samples, n_terms)
- Multivariate regressors.
-
- """
- # Check if BasisIndices is a sparse matrix
- sparsity = sp.sparse.issparse(basis_indices)
- if sparsity:
- basis_indices = basis_indices.toarray()
-
- # Initialization and consistency checks
- # number of input variables
- n_params = univ_p_val.shape[1]
-
- # Size of the experimental design
- n_samples = univ_p_val.shape[0]
-
- # number of basis terms
- n_terms = basis_indices.shape[0]
-
- # check that the variables have consistent sizes
- if n_params != basis_indices.shape[1]:
- raise ValueError(
- f"The shapes of basis_indices ({basis_indices.shape[1]}) and "
- f"univ_p_val ({n_params}) don't match!!"
- )
-
- # Preallocate the Psi matrix for performance
- psi = np.ones((n_samples, n_terms))
- # Assemble the Psi matrix
- for m in range(basis_indices.shape[1]):
- aa = np.where(basis_indices[:, m] > 0)[0]
- try:
- basisIdx = basis_indices[aa, m]
- bb = univ_p_val[:, m, basisIdx].reshape(psi[:, aa].shape)
- psi[:, aa] = np.multiply(psi[:, aa], bb)
- except ValueError as err:
- raise err
- return psi
# --------------------------------------------------------------------------------------------------------
- def adaptive_regression(self, ED_X, ED_Y, varIdx, verbose=False):
+ def adaptive_regression(self, ED_Y, varIdx, verbose=False):
"""
Adaptively fits the PCE model by comparing the scores of different
degrees and q-norm.
Parameters
----------
- ED_X : array of shape (n_samples, n_params)
- Experimental design.
ED_Y : array of shape (n_samples,)
Target values, i.e. simulation results for the Experimental design.
varIdx : int
@@ -743,7 +916,7 @@ class MetaModel():
"""
- n_samples, n_params = ED_X.shape
+ # n_samples, n_params = ED_X.shape
# Initialization
qAllCoeffs, AllCoeffs = {}, {}
qAllIndices_Sparse, AllIndices_Sparse = {}, {}
@@ -751,7 +924,7 @@ class MetaModel():
qAllnTerms, AllnTerms = {}, {}
qAllLCerror, AllLCerror = {}, {}
- # Extract degree array and qnorm array
+ # Extract degree array and q-norm array
deg_array = np.array([*self.allBasisIndices], dtype=int)
qnorm = [*self.allBasisIndices[str(int(deg_array[0]))]]
@@ -763,7 +936,7 @@ class MetaModel():
n_checks_qNorm = 2
nqnorms = len(qnorm)
qNormEarlyStop = True
- if nqnorms < n_checks_qNorm+1:
+ if nqnorms < n_checks_qNorm + 1:
qNormEarlyStop = False
# =====================================================================
@@ -771,6 +944,7 @@ class MetaModel():
# polynomial degree until the highest accuracy is reached
# =====================================================================
# For each degree check all q-norms and choose the best one
+ best_q = None
scores = -np.inf * np.ones(deg_array.shape[0])
qNormScores = -np.inf * np.ones(nqnorms)
@@ -783,38 +957,38 @@ class MetaModel():
BasisIndices = self.allBasisIndices[str(deg)][str(q)]
# Assemble the Psi matrix
- Psi = self.create_psi(BasisIndices, self.univ_p_val)
+ Psi = create_psi(BasisIndices, self.univ_p_val)
- # Calulate the cofficients of the meta model
+ # Calulate the cofficients of the metamodel
outs = self.regression(Psi, ED_Y, BasisIndices)
# Calculate and save the score of LOOCV
- score, LCerror = self.corr_loocv_error(outs['clf_poly'],
- outs['sparePsi'],
- outs['coeffs'],
- ED_Y)
+ score, LCerror = corr_loocv_error(outs['clf_poly'],
+ outs['sparePsi'],
+ outs['coeffs'],
+ ED_Y)
# Check the convergence of noise for FastARD
if self.pce_reg_method == 'FastARD' and \
- outs['clf_poly'].alpha_ < np.finfo(np.float32).eps:
+ outs['clf_poly'].alpha_ < np.finfo(np.float32).eps:
score = -np.inf
qNormScores[qidx] = score
- qAllCoeffs[str(qidx+1)] = outs['coeffs']
- qAllIndices_Sparse[str(qidx+1)] = outs['spareMulti-Index']
- qAllclf_poly[str(qidx+1)] = outs['clf_poly']
- qAllnTerms[str(qidx+1)] = BasisIndices.shape[0]
- qAllLCerror[str(qidx+1)] = LCerror
+ qAllCoeffs[str(qidx + 1)] = outs['coeffs']
+ qAllIndices_Sparse[str(qidx + 1)] = outs['spareMulti-Index']
+ qAllclf_poly[str(qidx + 1)] = outs['clf_poly']
+ qAllnTerms[str(qidx + 1)] = BasisIndices.shape[0]
+ qAllLCerror[str(qidx + 1)] = LCerror
# EarlyStop check
# if there are at least n_checks_qNorm entries after the
# best one, we stop
if qNormEarlyStop and \
- sum(np.isfinite(qNormScores)) > n_checks_qNorm:
+ sum(np.isfinite(qNormScores)) > n_checks_qNorm:
# If the error has increased the last two iterations, stop!
qNormScores_nonInf = qNormScores[np.isfinite(qNormScores)]
deltas = np.sign(np.diff(qNormScores_nonInf))
- if sum(deltas[-n_checks_qNorm+1:]) == 2:
+ if sum(deltas[-n_checks_qNorm + 1:]) == 2:
# stop the q-norm loop here
break
if np.var(ED_Y) == 0:
@@ -824,11 +998,11 @@ class MetaModel():
best_q = np.nanargmax(qNormScores)
scores[degIdx] = qNormScores[best_q]
- AllCoeffs[str(degIdx+1)] = qAllCoeffs[str(best_q+1)]
- AllIndices_Sparse[str(degIdx+1)] = qAllIndices_Sparse[str(best_q+1)]
- Allclf_poly[str(degIdx+1)] = qAllclf_poly[str(best_q+1)]
- AllnTerms[str(degIdx+1)] = qAllnTerms[str(best_q+1)]
- AllLCerror[str(degIdx+1)] = qAllLCerror[str(best_q+1)]
+ AllCoeffs[str(degIdx + 1)] = qAllCoeffs[str(best_q + 1)]
+ AllIndices_Sparse[str(degIdx + 1)] = qAllIndices_Sparse[str(best_q + 1)]
+ Allclf_poly[str(degIdx + 1)] = qAllclf_poly[str(best_q + 1)]
+ AllnTerms[str(degIdx + 1)] = qAllnTerms[str(best_q + 1)]
+ AllLCerror[str(degIdx + 1)] = qAllLCerror[str(best_q + 1)]
# Check the direction of the error (on average):
# if it increases consistently stop the iterations
@@ -836,7 +1010,7 @@ class MetaModel():
scores_nonInf = scores[scores != -np.inf]
ss = np.sign(scores_nonInf - np.max(scores_nonInf))
# ss<0 error decreasing
- errorIncreases = np.sum(np.sum(ss[-2:])) <= -1*n_checks_degree
+ errorIncreases = np.sum(np.sum(ss[-2:])) <= -1 * n_checks_degree
if errorIncreases:
break
@@ -847,7 +1021,7 @@ class MetaModel():
# ------------------ Summary of results ------------------
# Select the one with the best score and save the necessary outputs
- best_deg = np.nanargmax(scores)+1
+ best_deg = np.nanargmax(scores) + 1
coeffs = AllCoeffs[str(best_deg)]
basis_indices = AllIndices_Sparse[str(best_deg)]
clf_poly = Allclf_poly[str(best_deg)]
@@ -863,24 +1037,24 @@ class MetaModel():
nnz_idx = np.nonzero(coeffs)[0]
BasisIndices_Sparse = basis_indices[nnz_idx]
- print(f'Output variable {varIdx+1}:')
+ print(f'Output variable {varIdx + 1}:')
print('The estimation of PCE coefficients converged at polynomial '
- f'degree {deg_array[best_deg-1]} with '
+ f'degree {deg_array[best_deg - 1]} with '
f'{len(BasisIndices_Sparse)} terms (Sparsity index = '
- f'{round(len(BasisIndices_Sparse)/P, 3)}).')
+ f'{round(len(BasisIndices_Sparse) / P, 3)}).')
- print(f'Final ModLOO error estimate: {1-max(scores):.3e}')
- print('\n'+'-'*50)
+ print(f'Final ModLOO error estimate: {1 - max(scores):.3e}')
+ print('\n' + '-' * 50)
if verbose:
- print('='*50)
- print(' '*10 + ' Summary of results ')
- print('='*50)
+ print('=' * 50)
+ print(' ' * 10 + ' Summary of results ')
+ print('=' * 50)
print("Scores:\n", scores)
- print("Degree of best score:", self.deg_array[best_deg-1])
+ print("Degree of best score:", self.deg_array[best_deg - 1])
print("No. of terms:", len(basis_indices))
- print("Sparsity index:", round(len(basis_indices)/P, 3))
+ print("Sparsity index:", round(len(basis_indices) / P, 3))
print("Best Indices:\n", basis_indices)
if self.pce_reg_method in ['BRR', 'ARD']:
@@ -899,7 +1073,7 @@ class MetaModel():
plt.text(0.75, 0.5, text, fontsize=18, transform=ax.transAxes)
plt.show()
- print('='*80)
+ print('=' * 80)
# Create a dict to pass the outputs
returnVars = dict()
@@ -913,118 +1087,6 @@ class MetaModel():
return returnVars
- # -------------------------------------------------------------------------
- def corr_loocv_error(self, clf, psi, coeffs, y):
- """
- Calculates the corrected LOO error for regression on regressor
- matrix `psi` that generated the coefficients based on [1] and [2].
-
- [1] Blatman, G., 2009. Adaptive sparse polynomial chaos expansions for
- uncertainty propagation and sensitivity analysis (Doctoral
- dissertation, Clermont-Ferrand 2).
-
- [2] Blatman, G. and Sudret, B., 2011. Adaptive sparse polynomial chaos
- expansion based on least angle regression. Journal of computational
- Physics, 230(6), pp.2345-2367.
-
- Parameters
- ----------
- clf : object
- Fitted estimator.
- psi : array of shape (n_samples, n_features)
- The multivariate orthogonal polynomials (regressor).
- coeffs : array-like of shape (n_features,)
- Estimated cofficients.
- y : array of shape (n_samples,)
- Target values.
-
- Returns
- -------
- R_2 : float
- LOOCV Validation score (1-LOOCV erro).
- residual : array of shape (n_samples,)
- Residual values (y - predicted targets).
-
- """
- psi = np.array(psi, dtype=float)
-
- # Create PSI_Sparse by removing redundent terms
- nnz_idx = np.nonzero(coeffs)[0]
- if len(nnz_idx) == 0:
- nnz_idx = [0]
- psi_sparse = psi[:, nnz_idx]
-
- # NrCoeffs of aPCEs
- P = len(nnz_idx)
- # NrEvaluation (Size of experimental design)
- N = psi.shape[0]
-
- # Build the projection matrix
- PsiTPsi = np.dot(psi_sparse.T, psi_sparse)
-
- if np.linalg.cond(PsiTPsi) > 1e-12: #and \
- # np.linalg.cond(PsiTPsi) < 1/sys.float_info.epsilon:
- # faster
- try:
- M = sp.linalg.solve(PsiTPsi,
- sp.sparse.eye(PsiTPsi.shape[0]).toarray())
- except:
- raise AttributeError('There are too few samples for the corrected loo-cv error. Fit surrogate on at least as many samples as parameters to use this')
- else:
- # stabler
- M = np.linalg.pinv(PsiTPsi)
-
- # h factor (the full matrix is not calculated explicitly,
- # only the trace is, to save memory)
- PsiM = np.dot(psi_sparse, M)
-
- h = np.sum(np.multiply(PsiM, psi_sparse), axis=1, dtype=np.longdouble)#float128)
-
- # ------ Calculate Error Loocv for each measurement point ----
- # Residuals
- try:
- residual = clf.predict(psi) - y
- except:
- residual = np.dot(psi, coeffs) - y
-
- # Variance
- var_y = np.var(y)
-
- if var_y == 0:
- norm_emp_error = 0
- loo_error = 0
- LCerror = np.zeros((y.shape))
- return 1-loo_error, LCerror
- else:
- norm_emp_error = np.mean(residual**2)/var_y
-
- # LCerror = np.divide(residual, (1-h))
- LCerror = residual / (1-h)
- loo_error = np.mean(np.square(LCerror)) / var_y
- # if there are NaNs, just return an infinite LOO error (this
- # happens, e.g., when a strongly underdetermined problem is solved)
- if np.isnan(loo_error):
- loo_error = np.inf
-
- # Corrected Error for over-determined system
- tr_M = np.trace(M)
- if tr_M < 0 or abs(tr_M) > 1e6:
- tr_M = np.trace(np.linalg.pinv(np.dot(psi.T, psi)))
-
- # Over-determined system of Equation
- if N > P:
- T_factor = N/(N-P) * (1 + tr_M)
-
- # Under-determined system of Equation
- else:
- T_factor = np.inf
-
- corrected_loo_error = loo_error * T_factor
-
- R_2 = 1 - corrected_loo_error
-
- return R_2, LCerror
-
# -------------------------------------------------------------------------
def pca_transformation(self, target, verbose=False):
"""
@@ -1034,6 +1096,9 @@ class MetaModel():
----------
target : array of shape (n_samples,)
Target values.
+ verbose : bool
+ Set to True to get more information during functtion call.
+ The default is False.
Returns
-------
@@ -1046,20 +1111,20 @@ class MetaModel():
"""
# Transform via Principal Component Analysis
- if hasattr(self, 'var_pca_threshold'):
+ if self.var_pca_threshold is None:
var_pca_threshold = self.var_pca_threshold
else:
var_pca_threshold = 100.0
n_samples, n_features = target.shape
- if hasattr(self, 'n_pca_components'):
+ if self.n_pca_components is None:
n_pca_components = self.n_pca_components
else:
# Instantiate and fit sklearnPCA object
covar_matrix = sklearnPCA(n_components=None)
covar_matrix.fit(target)
var = np.cumsum(np.round(covar_matrix.explained_variance_ratio_,
- decimals=5)*100)
+ decimals=5) * 100)
# Find the number of components to explain self.varPCAThreshold of
# variance
try:
@@ -1084,95 +1149,16 @@ class MetaModel():
return pca, scaled_target, n_pca_components
- # -------------------------------------------------------------------------
- def gaussian_process_emulator(self, X, y, nug_term=None, autoSelect=False,
- varIdx=None):
- """
- Fits a Gaussian Process Emulator to the target given the training
- points.
-
- Parameters
- ----------
- X : array of shape (n_samples, n_params)
- Training points.
- y : array of shape (n_samples,)
- Target values.
- nug_term : float, optional
- Nugget term. The default is None, i.e. variance of y.
- autoSelect : bool, optional
- Loop over some kernels and select the best. The default is False.
- varIdx : int, optional
- The index number. The default is None.
-
- Returns
- -------
- gp : object
- Fitted estimator.
-
- """
-
- nug_term = nug_term if nug_term else np.var(y)
-
- Kernels = [nug_term * kernels.RBF(length_scale=1.0,
- length_scale_bounds=(1e-25, 1e15)),
- nug_term * kernels.RationalQuadratic(length_scale=0.2,
- alpha=1.0),
- nug_term * kernels.Matern(length_scale=1.0,
- length_scale_bounds=(1e-15, 1e5),
- nu=1.5)]
-
- # Automatic selection of the kernel
- if autoSelect:
- gp = {}
- BME = []
- for i, kernel in enumerate(Kernels):
- gp[i] = GaussianProcessRegressor(kernel=kernel,
- n_restarts_optimizer=3,
- normalize_y=False)
-
- # Fit to data using Maximum Likelihood Estimation
- gp[i].fit(X, y)
-
- # Store the MLE as BME score
- BME.append(gp[i].log_marginal_likelihood())
-
- gp = gp[np.argmax(BME)]
-
- else:
- gp = GaussianProcessRegressor(kernel=Kernels[0],
- n_restarts_optimizer=3,
- normalize_y=False)
- gp.fit(X, y)
-
- # Compute score
- if varIdx is not None:
- Score = gp.score(X, y)
- print('-'*50)
- print(f'Output variable {varIdx}:')
- print('The estimation of GPE coefficients converged,')
- print(f'with the R^2 score: {Score:.3f}')
- print('-'*50)
-
- return gp
-
# -------------------------------------------------------------------------
def eval_metamodel(self, samples):
"""
- Evaluates meta-model at the requested samples. One can also generate
+ Evaluates metamodel at the requested samples. One can also generate
nsamples.
Parameters
----------
samples : array of shape (n_samples, n_params), optional
- Samples to evaluate meta-model at. The default is None.
- nsamples : int, optional
- Number of samples to generate, if no `samples` is provided. The
- default is None.
- sampling_method : str, optional
- Type of sampling, if no `samples` is provided. The default is
- 'random'.
- return_samples : bool, optional
- Retun samples, if no `samples` is provided. The default is False.
+ Samples to evaluate metamodel at. The default is None.
Returns
-------
@@ -1183,29 +1169,33 @@ class MetaModel():
"""
# Transform into np array - can also be given as list
samples = np.array(samples)
-
+
# Transform samples to the independent space
samples = self.InputSpace.transform(
samples,
method='user'
- )
+ )
# Compute univariate bases for the given samples
+ univ_p_val = None
if self.meta_model_type.lower() != 'gpe':
univ_p_val = self.univ_basis_vals(
samples,
n_max=np.max(self.pce_deg)
- )
+ )
+ mean_pred = None
+ std_pred = None
mean_pred_b = {}
std_pred_b = {}
+ b_i = 0
# Loop over bootstrap iterations
for b_i in range(self.n_bootstrap_itrs):
# Extract model dictionary
if self.meta_model_type.lower() == 'gpe':
- model_dict = self.gp_poly[f'b_{b_i+1}']
+ model_dict = self.gp_poly[f'b_{b_i + 1}']
else:
- model_dict = self.coeffs_dict[f'b_{b_i+1}']
+ model_dict = self.coeffs_dict[f'b_{b_i + 1}']
# Loop over outputs
mean_pred = {}
@@ -1219,30 +1209,30 @@ class MetaModel():
# Prediction with GPE
if self.meta_model_type.lower() == 'gpe':
- X_T = self.x_scaler[f'b_{b_i+1}'][output].transform(samples)
- gp = self.gp_poly[f'b_{b_i+1}'][output][in_key]
+ X_T = self.x_scaler[f'b_{b_i + 1}'][output].transform(samples)
+ gp = self.gp_poly[f'b_{b_i + 1}'][output][in_key]
y_mean, y_std = gp.predict(X_T, return_std=True)
else:
# Prediction with PCE
# Assemble Psi matrix
- basis = self.basis_dict[f'b_{b_i+1}'][output][in_key]
- psi = self.create_psi(basis, univ_p_val)
+ basis = self.basis_dict[f'b_{b_i + 1}'][output][in_key]
+ psi = create_psi(basis, univ_p_val)
# Prediction
if self.bootstrap_method != 'fast' or b_i == 0:
# with error bar, i.e. use clf_poly
- clf_poly = self.clf_poly[f'b_{b_i+1}'][output][in_key]
+ clf_poly = self.clf_poly[f'b_{b_i + 1}'][output][in_key]
try:
y_mean, y_std = clf_poly.predict(
psi, return_std=True
- )
+ )
except TypeError:
y_mean = clf_poly.predict(psi)
y_std = np.zeros_like(y_mean)
else:
# without error bar
- coeffs = self.coeffs_dict[f'b_{b_i+1}'][output][in_key]
+ coeffs = self.coeffs_dict[f'b_{b_i + 1}'][output][in_key]
y_mean = np.dot(psi, coeffs)
y_std = np.zeros_like(y_mean)
@@ -1252,7 +1242,7 @@ class MetaModel():
# Save predictions for each output
if self.dim_red_method.lower() == 'pca':
- PCA = self.pca[f'b_{b_i+1}'][output]
+ PCA = self.pca[f'b_{b_i + 1}'][output]
mean_pred[output] = PCA.inverse_transform(mean)
std_pred[output] = np.zeros(mean.shape)
else:
@@ -1283,12 +1273,13 @@ class MetaModel():
if self.n_bootstrap_itrs > 1:
std_pred[output] = np.std(outs, axis=0)
else:
+ # TODO: this b_i seems off here
std_pred[output] = std_pred_b[b_i][output]
return mean_pred, std_pred
# -------------------------------------------------------------------------
- def create_model_error(self, X, y, Model, name='Calib'):
+ def create_model_error(self, X, y, MeasuredData):
"""
Fits a GPE-based model error.
@@ -1299,8 +1290,7 @@ class MetaModel():
extracted data.
y : array of shape (n_outputs,)
The model response for the MAP parameter set.
- name : str, optional
- Calibration or validation. The default is `'Calib'`.
+ MeasuredData :
Returns
-------
@@ -1315,7 +1305,7 @@ class MetaModel():
# Read data
# TODO: do this call outside the metamodel
- MeasuredData = Model.read_observation(case=name)
+ # MeasuredData = Model.read_observation(case=name)
# Fitting GPR based bias model
for out in outputNames:
@@ -1331,7 +1321,7 @@ class MetaModel():
delta = data # - y[out][0]
BiasInputs = np.hstack((X[out], y[out].reshape(-1, 1)))
X_S = scaler.fit_transform(BiasInputs)
- gp = self.gaussian_process_emulator(X_S, delta)
+ gp = gaussian_process_emulator(X_S, delta)
self.errorScale[out]["y_1"] = scaler
self.errorclf_poly[out]["y_1"] = gp
@@ -1413,13 +1403,13 @@ class MetaModel():
"""
# TODO: what properties should be moved to the new object?
new_MetaModelOpts = copy.deepcopy(self)
- new_MetaModelOpts.input_obj = self.input_obj#InputObj
+ new_MetaModelOpts.input_obj = self.input_obj # InputObj
new_MetaModelOpts.InputSpace = self.InputSpace
- #new_MetaModelOpts.InputSpace.meta_Model = 'aPCE'
- #new_MetaModelOpts.InputSpace.InputObj = self.input_obj
- #new_MetaModelOpts.InputSpace.ndim = len(self.input_obj.Marginals)
+ # new_MetaModelOpts.InputSpace.meta_Model = 'aPCE'
+ # new_MetaModelOpts.InputSpace.InputObj = self.input_obj
+ # new_MetaModelOpts.InputSpace.ndim = len(self.input_obj.Marginals)
new_MetaModelOpts.n_params = len(self.input_obj.Marginals)
- #new_MetaModelOpts.InputSpace.hdf5_file = None
+ # new_MetaModelOpts.InputSpace.hdf5_file = None
return new_MetaModelOpts
@@ -1439,15 +1429,15 @@ class MetaModel():
Returns
-------
deg_array: array
- Array containing the arrays.
+ The selected degrees.
"""
# Define the deg_array
max_deg = np.max(self.pce_deg)
min_Deg = np.min(self.pce_deg)
-
+
# TODO: remove the options for sequential?
- #nitr = n_samples - self.InputSpace.n_init_samples
+ nitr = n_samples - self.InputSpace.n_init_samples
# Check q-norm
if not np.isscalar(self.pce_q_norm):
@@ -1455,49 +1445,61 @@ class MetaModel():
else:
self.pce_q_norm = np.array([self.pce_q_norm])
- def M_uptoMax(maxDeg):
- n_combo = np.zeros(maxDeg)
- for i, d in enumerate(range(1, maxDeg+1)):
- n_combo[i] = math.factorial(ndim+d)
- n_combo[i] /= math.factorial(ndim) * math.factorial(d)
- return n_combo
+ # def M_uptoMax(maxDeg):
+ # n_combo = np.zeros(maxDeg)
+ # for i, d in enumerate(range(1, maxDeg + 1)):
+ # n_combo[i] = math.factorial(ndim + d)
+ # n_combo[i] /= math.factorial(ndim) * math.factorial(d)
+ # return n_combo
deg_new = max_deg
- #d = nitr if nitr != 0 and self.n_params > 5 else 1
+ # d = nitr if nitr != 0 and self.n_params > 5 else 1
# d = 1
# min_index = np.argmin(abs(M_uptoMax(max_deg)-ndim*n_samples*d))
# deg_new = range(1, max_deg+1)[min_index]
if deg_new > min_Deg and self.pce_reg_method.lower() != 'fastard':
- deg_array = np.arange(min_Deg, deg_new+1)
+ deg_array = np.arange(min_Deg, deg_new + 1)
else:
deg_array = np.array([deg_new])
return deg_array
def generate_polynomials(self, max_deg=None):
+ """
+ Generates (univariate) polynomials.
+
+ Parameters
+ ----------
+ max_deg : int
+ Maximum polynomial degree.
+
+ Returns
+ -------
+ None
+ """
# Check for InputSpace
- if not hasattr(self, 'InputSpace'):
+ if self.InputSpace is None:
raise AttributeError('Generate or add InputSpace before generating polynomials')
-
+
ndim = self.InputSpace.ndim
# Create orthogonal polynomial coefficients if necessary
- if (self.meta_model_type.lower()!='gpe') and max_deg is not None:# and self.input_obj.poly_coeffs_flag:
+ if (self.meta_model_type.lower() != 'gpe') and max_deg is not None: # and self.input_obj.poly_coeffs_flag:
self.polycoeffs = {}
for parIdx in tqdm(range(ndim), ascii=True,
desc="Computing orth. polynomial coeffs"):
poly_coeffs = apoly_construction(
self.InputSpace.raw_data[parIdx],
max_deg
- )
- self.polycoeffs[f'p_{parIdx+1}'] = poly_coeffs
+ )
+ self.polycoeffs[f'p_{parIdx + 1}'] = poly_coeffs
else:
raise AttributeError('MetaModel cannot generate polynomials in the given scenario!')
# -------------------------------------------------------------------------
def _compute_pce_moments(self):
"""
- Computes the first two moments using the PCE-based meta-model.
+ Computes the first two moments using the PCE-based metamodel.
Returns
-------
@@ -1507,11 +1509,11 @@ class MetaModel():
The second moment (standard deviation) of the surrogate.
"""
-
- # Check if its truly a pce-surrogate
+
+ # Check if it's truly a pce-surrogate
if self.meta_model_type.lower() == 'gpe':
raise AttributeError('Moments can only be computed for pce-type surrogates')
-
+
outputs = self.out_names
pce_means_b = {}
pce_stds_b = {}
@@ -1519,7 +1521,7 @@ class MetaModel():
# Loop over bootstrap iterations
for b_i in range(self.n_bootstrap_itrs):
# Loop over the metamodels
- coeffs_dicts = self.coeffs_dict[f'b_{b_i+1}'].items()
+ coeffs_dicts = self.coeffs_dict[f'b_{b_i + 1}'].items()
means = {}
stds = {}
for output, coef_dict in coeffs_dicts:
@@ -1529,20 +1531,20 @@ class MetaModel():
for index, values in coef_dict.items():
idx = int(index.split('_')[1]) - 1
- coeffs = self.coeffs_dict[f'b_{b_i+1}'][output][index]
+ coeffs = self.coeffs_dict[f'b_{b_i + 1}'][output][index]
# Mean = c_0
if coeffs[0] != 0:
pce_mean[idx] = coeffs[0]
else:
- clf_poly = self.clf_poly[f'b_{b_i+1}'][output]
+ clf_poly = self.clf_poly[f'b_{b_i + 1}'][output]
pce_mean[idx] = clf_poly[index].intercept_
# Var = sum(coeffs[1:]**2)
pce_var[idx] = np.sum(np.square(coeffs[1:]))
# Save predictions for each output
if self.dim_red_method.lower() == 'pca':
- PCA = self.pca[f'b_{b_i+1}'][output]
+ PCA = self.pca[f'b_{b_i + 1}'][output]
means[output] = PCA.inverse_transform(pce_mean)
stds[output] = PCA.inverse_transform(np.sqrt(pce_var))
else:
diff --git a/tests/test_BayesInference.py b/tests/test_BayesInference.py
new file mode 100644
index 0000000000000000000000000000000000000000..2f22f9158190c930fe6827fe892e3a87027e33c5
--- /dev/null
+++ b/tests/test_BayesInference.py
@@ -0,0 +1,1105 @@
+# -*- coding: utf-8 -*-
+"""
+Test the BayesInference class for bayesvalidrox
+
+Tests are available for the following functions
+ _logpdf - x
+ _kernel_rbf - x
+class BayesInference:
+ setup_inference - x
+ create_inference - x
+ perform_bootstrap Need working model for tests without emulator
+ _perturb_data - x
+ create_error_model Error in the MetaModel
+ _eval_model Need working model to test this
+ normpdf - x
+ _corr_factor_BME_old - removed
+ _corr_factor_BME - x
+ _rejection_sampling - x
+ _posterior_predictive - x
+ plot_post_params - x
+ plot_log_BME - x
+ _plot_max_a_posteriori Need working model to test this
+ _plot_post_predictive - x
+"""
+import sys
+import pytest
+import numpy as np
+import pandas as pd
+
+from bayesvalidrox.surrogate_models.inputs import Input
+from bayesvalidrox.surrogate_models.exp_designs import ExpDesigns
+from bayesvalidrox.surrogate_models.surrogate_models import MetaModel
+from bayesvalidrox.pylink.pylink import PyLinkForwardModel as PL
+from bayesvalidrox.surrogate_models.engine import Engine
+from bayesvalidrox.bayes_inference.discrepancy import Discrepancy
+from bayesvalidrox.bayes_inference.mcmc import MCMC
+from bayesvalidrox.bayes_inference.bayes_inference import BayesInference
+from bayesvalidrox.bayes_inference.bayes_inference import _logpdf, _kernel_rbf
+
+sys.path.append("src/")
+sys.path.append("../src/")
+
+
+#%% Test _logpdf
+
+def test_logpdf() -> None:
+ """
+ Calculate loglikelihood
+
+ """
+ _logpdf([0], [0], [1])
+
+
+#%% Test _kernel_rbf
+
+def test_kernel_rbf() -> None:
+ """
+ Create RBF kernel
+ """
+ X = [[0, 0], [1, 1.5]]
+ pars = [1, 0.5, 1]
+ _kernel_rbf(X, pars)
+
+
+def test_kernel_rbf_lesspar() -> None:
+ """
+ Create RBF kernel with too few parameters
+ """
+ X = [[0, 0], [1, 1.5]]
+ pars = [1, 2]
+ with pytest.raises(AttributeError) as excinfo:
+ _kernel_rbf(X, pars)
+ assert str(excinfo.value) == 'Provide 3 parameters for the RBF kernel!'
+
+
+#%% Test MCMC init
+
+def test_BayesInference() -> None:
+ """
+ Construct a BayesInference object
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mod = PL()
+ mm = MetaModel(inp)
+ expdes = ExpDesigns(inp)
+ engine = Engine(mm, mod, expdes)
+ BayesInference(engine)
+
+
+#%% Test create_inference
+# TODO: disabled this test!
+def test_create_inference() -> None:
+ """
+ Run inference
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0]) # Error in plots if this is not available
+
+ mm = MetaModel(inp)
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])} # Error if x_values not given
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts # Error if this not class 'DiscrepancyOpts' or dict(?)
+ bi.bootstrap = True # Error if this and bayes_loocv and just_analysis are all False?
+ bi.plot_post_pred = False # Remaining issue in the violinplot
+ bi.create_inference()
+ # Remaining issue in the violinplot in plot_post_predictive
+
+
+#%% Test rejection_sampling
+def test_rejection_sampling_nologlik() -> None:
+ """
+ Perform rejection sampling without given log likelihood
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mod = PL()
+ mm = MetaModel(inp)
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ engine = Engine(mm, mod, expdes)
+ bi = BayesInference(engine)
+ bi.prior_samples = expdes.generate_samples(100, 'random')
+ with pytest.raises(AttributeError) as excinfo:
+ bi._rejection_sampling()
+ assert str(excinfo.value) == 'No log-likelihoods available!'
+
+
+def test_rejection_sampling_noprior() -> None:
+ """
+ Perform rejection sampling without prior samples
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mod = PL()
+ mm = MetaModel(inp)
+ expdes = ExpDesigns(inp)
+ engine = Engine(mm, mod, expdes)
+ bi = BayesInference(engine)
+ with pytest.raises(AttributeError) as excinfo:
+ bi._rejection_sampling()
+ assert str(excinfo.value) == 'No prior samples available!'
+
+
+def test_rejection_sampling() -> None:
+ """
+ Perform rejection sampling
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mod = PL()
+ mm = MetaModel(inp)
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ engine = Engine(mm, mod, expdes)
+ bi = BayesInference(engine)
+ bi.prior_samples = expdes.generate_samples(100, 'random')
+ bi.log_likes = np.swapaxes(np.atleast_2d(np.log(np.random.random(100) * 3)), 0, 1)
+ bi._rejection_sampling()
+
+
+#%% Test _perturb_data
+
+def test_perturb_data() -> None:
+ """
+ Perturb data
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mod = PL()
+ mm = MetaModel(inp)
+ expdes = ExpDesigns(inp)
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ data = pd.DataFrame()
+ data['Z'] = [0.45]
+ bi._perturb_data(data, ['Z'])
+
+
+def test_perturb_data_loocv() -> None:
+ """
+ Perturb data with bayes_loocv
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mod = PL()
+ mm = MetaModel(inp)
+ expdes = ExpDesigns(inp)
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ data = pd.DataFrame()
+ data['Z'] = [0.45]
+ bi.bayes_loocv = True
+ bi._perturb_data(data, ['Z'])
+
+
+#%% Test _eval_model
+
+def test_eval_model() -> None:
+ """
+ Run model with descriptive key
+ """
+ # TODO: need functioning example model to test this
+ None
+
+
+#%% Test corr_factor_BME
+
+def test_corr_factor_BME() -> None:
+ """
+ Calculate correction factor
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+
+ mm = MetaModel(inp)
+ mm.fit(expdes.X, expdes.Y)
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ obs_data = {'Z': np.array([0.45])}
+ total_sigma2s = {'Z': np.array([0.15])}
+ logBME = [0, 0, 0]
+
+ bi = BayesInference(engine)
+ bi.selected_indices = {'Z': 0}
+ bi._corr_factor_BME(obs_data, total_sigma2s, logBME)
+
+
+def test_corr_factor_BME_selectedindices() -> None:
+ """
+ Calculate correction factor
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+
+ mm = MetaModel(inp)
+ mm.fit(expdes.X, expdes.Y)
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ obs_data = {'Z': np.array([0.45])}
+ total_sigma2s = {'Z': np.array([0.15])}
+ logBME = [0, 0, 0]
+
+ bi = BayesInference(engine)
+ bi.selected_indices = {'Z': 0}
+ bi._corr_factor_BME(obs_data, total_sigma2s, logBME)
+
+
+#%% Test normpdf
+
+def test_normpdf_nosigmas() -> None:
+ """
+ Run normpdf without any additional sigmas
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': np.array([[0.4], [0.5], [0.45]])}
+
+ mm = MetaModel(inp)
+ mod = PL()
+ mod.Output.names = ['Z']
+ engine = Engine(mm, mod, expdes)
+
+ obs_data = {'Z': np.array([0.45])}
+ total_sigma2s = {'Z': np.array([0.15])}
+
+ bi = BayesInference(engine)
+ bi.normpdf(expdes.Y, obs_data, total_sigma2s, sigma2=None, std=None)
+
+
+def test_normpdf_sigma2() -> None:
+ """
+ Run normpdf with sigma2
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': np.array([[0.4], [0.5], [0.45]])}
+
+ mm = MetaModel(inp)
+ mod = PL()
+ mod.Output.names = ['Z']
+ engine = Engine(mm, mod, expdes)
+
+ obs_data = {'Z': np.array([0.45])}
+ total_sigma2s = {'Z': np.array([0.15])}
+ sigma2 = [[0]]
+
+ bi = BayesInference(engine)
+ bi.normpdf(expdes.Y, obs_data, total_sigma2s, sigma2=sigma2, std=None)
+
+
+def test_normpdf_allsigmas() -> None:
+ """
+ Run normpdf with all additional sigmas
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': np.array([[0.4], [0.5], [0.45]])}
+
+ mm = MetaModel(inp)
+ mod = PL()
+ mod.Output.names = ['Z']
+ engine = Engine(mm, mod, expdes)
+
+ obs_data = {'Z': np.array([0.45])}
+ total_sigma2s = {'Z': np.array([0.15])}
+ sigma2 = [[0]]
+
+ bi = BayesInference(engine)
+ bi.normpdf(expdes.Y, obs_data, total_sigma2s, sigma2=sigma2, std=total_sigma2s)
+
+
+#%% Test setup_inference
+
+def test_setup_inference_noobservation() -> None:
+ """
+ Test the object setup without given observations
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ with pytest.raises(Exception) as excinfo:
+ bi.setup_inference()
+ assert str(
+ excinfo.value) == ('Please provide the observation data as a dictionary via observations attribute or pass the '
+ 'csv-file path to MeasurementFile attribute')
+
+
+def test_setup_inference() -> None:
+ """
+ Test the object setup with observations
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.setup_inference()
+
+
+def test_setup_inference_priorsamples() -> None:
+ """
+ Test the object setup with prior samples set by hand
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.prior_samples = np.swapaxes(np.array([np.random.normal(0, 1, 100)]), 0, 1)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.setup_inference()
+
+
+def test_setup_inference_valid() -> None:
+ """
+ Test the object setup for valid
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations_valid = {'Z': np.array([0.45])}
+ mod.observations_valid = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.name = 'valid'
+ bi.setup_inference()
+
+
+def test_setup_inference_noname() -> None:
+ """
+ Test the object setup for an invalid inference name
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.name = ''
+ with pytest.raises(Exception) as excinfo:
+ bi.setup_inference()
+ assert str(excinfo.value) == 'The set inference type is not known! Use either `calib` or `valid`'
+
+
+#%% Test perform_bootstrap
+
+def test_perform_bootstrap() -> None:
+ """
+ Do bootstrap
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0]) # Error in plots if this is not available
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])} # Error if x_values not given
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.bootstrap = True
+ bi.plot_post_pred = False
+ total_sigma2s = {'Z': np.array([0.15])}
+ bi.setup_inference()
+ bi.perform_bootstrap(total_sigma2s)
+
+
+def test_perform_bootstrap_bayesloocv() -> None:
+ """
+ Do bootstrap
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0]) # Error in plots if this is not available
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])} # Error if x_values not given
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.bootstrap = True
+ bi.plot_post_pred = False
+ total_sigma2s = {'Z': np.array([0.15])}
+ bi.setup_inference()
+ bi.bayes_loocv = True
+ bi.perform_bootstrap(total_sigma2s)
+
+
+#%% Test create_error_model
+
+def create_error_model_prior() -> None:
+ """
+ Test creating MetaModel error-model for 'prior'
+ """
+ # TODO: there are issues with the expected formats from the MetaModel
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.bootstrap = True
+ bi.setup_inference()
+ bi.bias_inputs = expdes.X
+ bi.create_error_model(type_='prior', opt_sigma='B', sampler=None)
+
+
+def create_error_model_posterior() -> None:
+ """
+ Test creating MetaModel error-model for 'posterior'
+ """
+ # TODO: there are issues with the expected formats from the MetaModel
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ posterior = pd.DataFrame()
+ posterior[None] = [0, 1, 0.5]
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.bootstrap = True
+ bi.setup_inference()
+ bi.bias_inputs = expdes.X
+ bi.posterior_df = posterior
+ bi.create_error_model(type_='posterior', opt_sigma='B', sampler=None)
+
+
+#%% Test _posterior_predictive
+
+def test_posterior_predictive() -> None:
+ """
+ Test posterior predictions
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ prior_samples = np.swapaxes(np.array([np.random.normal(0, 1, 10)]), 0, 1)
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0]) # Error in plots if this is not available
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+ y_hat, y_std = mm.eval_metamodel(prior_samples)
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])} # Error if x_values not given
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ total_sigma2s = {'Z': np.array([0.15])}
+ posterior = pd.DataFrame()
+ posterior[None] = [0, 1, 0.5]
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.bootstrap = True
+ bi.plot_post_pred = False
+ bi.posterior_df = posterior
+ bi.bias_inputs = expdes.X
+ bi._mean_pce_prior_pred = y_hat
+ bi._std_pce_prior_pred = y_std
+ bi.Discrepancy.total_sigma2 = total_sigma2s
+ bi.setup_inference()
+ bi._posterior_predictive()
+
+
+def test_posterior_predictive_rejection() -> None:
+ """
+ Test posterior predictions with rejection inference
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ prior_samples = np.swapaxes(np.array([np.random.normal(0, 1, 10)]), 0, 1)
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0]) # Error in plots if this is not available
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+ y_hat, y_std = mm.eval_metamodel(prior_samples)
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])} # Error if x_values not given
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ total_sigma2s = {'Z': np.array([0.15])}
+ posterior = pd.DataFrame()
+ posterior[None] = [0, 1, 0.5]
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts
+ bi.bootstrap = True
+ bi.plot_post_pred = False
+ bi.posterior_df = posterior
+ bi.bias_inputs = expdes.X
+ bi._mean_pce_prior_pred = y_hat
+ bi._std_pce_prior_pred = y_std
+ bi.Discrepancy.total_sigma2 = total_sigma2s
+ bi.inference_method = 'rejection'
+ bi.setup_inference()
+ bi._posterior_predictive()
+
+
+#%% Test plot_post_params
+
+def test_plot_post_params() -> None:
+ """
+ Plot posterior dist
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ posterior = pd.DataFrame()
+ posterior[None] = [0, 1, 0.5]
+ bi.posterior_df = posterior
+ bi.plot_post_params('B')
+
+
+def test_plot_post_params_noemulator() -> None:
+ """
+ Plot posterior dist with emulator = False
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ posterior = pd.DataFrame()
+ posterior[None] = [0, 1, 0.5]
+ bi.posterior_df = posterior
+ bi.emulator = False
+ bi.plot_post_params('B')
+
+
+#%% Test plot_log_BME
+
+def test_plot_log_BME() -> None:
+ """
+ Show the log_BME from bootstrapping
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ bi.log_BME = np.array([[0, 0.2], [0, 0.2]])
+ bi.n_tot_measurement = 1
+ bi.plot_log_BME()
+
+
+def test_plot_log_BME_noemulator() -> None:
+ """
+ Show the log_BME from bootstrapping with emulator = False
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ bi.log_BME = np.array([[0, 0.2], [0, 0.2]])
+ bi.n_tot_measurement = 1
+ bi.emulator = False
+ bi.plot_log_BME()
+
+
+#%% Test _plot_max_a_posteriori
+
+def test_plot_max_a_posteriori_rejection() -> None:
+ """
+ Plot MAP estimate for rejection
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ bi.inference_method = 'rejection'
+ bi._plot_post_predictive()
+
+
+def test_plot_max_a_posteriori() -> None:
+ """
+ Plot MAP estimate
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ bi._plot_post_predictive()
+
+
+#%% Test _plot_post_predictive
+
+
+def test_plot_post_predictive_rejection() -> None:
+ """
+ Plot posterior predictions for rejection
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ bi.inference_method = 'rejection'
+ bi._plot_post_predictive()
+
+
+def test_plot_post_predictive() -> None:
+ """
+ Plot posterior predictions
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mod = PL()
+ engine = Engine(mm, mod, expdes)
+
+ bi = BayesInference(engine)
+ bi._plot_post_predictive()
+
+
+#%% Main runs
+if __name__ == '__main__':
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ # prior_samples = np.swapaxes(np.array([np.random.normal(0,1,10)]),0,1)
+
+ expdes = ExpDesigns(inp)
+ expdes.init_param_space(max_deg=1)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0]) # Error in plots if this is not
+
+ mm = MetaModel(inp)
+ mm.n_params = 1
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(1))
+ # y_hat, y_std = mm.eval_metamodel(prior_samples)
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45])}
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])} # Error if x_values not given
+ mod.Output.names = ['Z']
+ mod.n_obs = 1
+
+ engine = Engine(mm, mod, expdes)
+
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ obsData = pd.DataFrame({'Z': np.array([0.45]), 'x_values': np.array([0])}, columns=mod.Output.names)
+ DiscrepancyOpts = Discrepancy('')
+ DiscrepancyOpts.type = 'Gaussian'
+ DiscrepancyOpts.parameters = (obsData * 0.15) ** 2
+ DiscrepancyOpts.opt_sigma = 'B'
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = DiscrepancyOpts # Error if this not class 'DiscrepancyOpts' or dict(?)
+ bi.bootstrap = True # Error if this and bayes_loocv and just_analysis are all False?
+ bi.plot_post_pred = False # Remaining issue in the violinplot
+ bi.error_model = False
+ bi.bayes_loocv = True
+ if 1:
+ bi.create_inference()
+ # opt_sigma = 'B'
+ # total_sigma2s = {'Z':np.array([0.15])}
+ # data = pd.DataFrame()
+ # data['Z'] = [0.45]
+ # data['x_values'] = [0.3]
+ # bi.setup_inference()
+ # bi.perform_bootstrap(total_sigma2s)
+ posterior = pd.DataFrame()
+ posterior[None] = [0, 1, 0.5]
+ bi.posterior_df = posterior
+ # bi.bias_inputs = expdes.X
+ # bi._mean_pce_prior_pred = y_hat
+ # bi._std_pce_prior_pred = y_std
+ # bi.Discrepancy.total_sigma2 = total_sigma2s
+ # bi.create_error_model(type_ = 'posterior', opt_sigma = 'B', sampler = None)
+ # bi._posterior_predictive()
+ # bi.plot_post_params('B')
+ # bi.log_BME = np.array([[0,0.2],[0,0.2]])
+ # bi.n_tot_measurement = 1
+ # bi.plot_log_BME()
+ bi.inference_method = 'rejection'
+ bi._plot_max_a_posteriori()
diff --git a/tests/test_BayesModelComparison.py b/tests/test_BayesModelComparison.py
new file mode 100644
index 0000000000000000000000000000000000000000..91f328ec7ae39cf7cded6b228edc1442053a2dfb
--- /dev/null
+++ b/tests/test_BayesModelComparison.py
@@ -0,0 +1,28 @@
+# -*- coding: utf-8 -*-
+"""
+Test the BayesModelComparison class in bayesvalidrox.
+Tests are available for the following functions
+Class BayesModelComparison:
+ create_model_comparison
+ compare_models
+ generate_dataset
+ __perturb_data
+ cal_model_weight
+ plot_just_analysis
+ plot_model_weights
+ plot_bayes_factor
+
+"""
+import sys
+sys.path.append("src/")
+import pytest
+import numpy as np
+
+from bayesvalidrox.bayes_inference.bayes_model_comparison import BayesModelComparison
+#from bayesvalidrox.surrogate_models.input_space import InputSpace
+
+def test_BMC() -> None:
+ """
+ Build BMC without inputs
+ """
+ BayesModelComparison()
\ No newline at end of file
diff --git a/tests/test_Discrepancy.py b/tests/test_Discrepancy.py
index c46e0a13751756e0583f3176489e7215da77f4ba..7fb948d905031e7d7e6235857c27792fb29ece57 100644
--- a/tests/test_Discrepancy.py
+++ b/tests/test_Discrepancy.py
@@ -36,22 +36,8 @@ def test_get_sample() -> None:
"""
Get discrepancy sample
"""
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- disc = Discrepancy(InputDisc = inp)
+ disc = Discrepancy()
with pytest.raises(AttributeError) as excinfo:
disc.get_sample(2)
assert str(excinfo.value) == 'Cannot create new samples, please provide input distributions'
-
-
-
-
-if __name__ == '__main__':
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- disc = Discrepancy(InputDisc = inp)
- disc.get_sample(2)
\ No newline at end of file
+
\ No newline at end of file
diff --git a/tests/test_engine.py b/tests/test_Engine.py
similarity index 56%
rename from tests/test_engine.py
rename to tests/test_Engine.py
index 72dabd466174fbfd2b60ed23fc500e53ad51f240..6b03a26237d1b1ff741bcfd69e8ffff188cc8d7d 100644
--- a/tests/test_engine.py
+++ b/tests/test_Engine.py
@@ -31,10 +31,7 @@ Engine:
import math
import numpy as np
import pandas as pd
-
import sys
-sys.path.append("src/")
-#import pytest
from bayesvalidrox.surrogate_models.inputs import Input
from bayesvalidrox.surrogate_models.exp_designs import ExpDesigns
@@ -43,6 +40,8 @@ from bayesvalidrox.pylink.pylink import PyLinkForwardModel as PL
from bayesvalidrox.surrogate_models.engine import Engine
from bayesvalidrox.surrogate_models.engine import hellinger_distance, logpdf, subdomain
+sys.path.append("src/")
+
#%% Test Engine constructor
@@ -54,12 +53,13 @@ def test_engine() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mod = PL()
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
Engine(mm, mod, expdes)
+
#%% Test Engine.start_engine
def test_start_engine() -> None:
@@ -69,7 +69,7 @@ def test_start_engine() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mod = PL()
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
@@ -89,9 +89,9 @@ def test__error_Mean_Std() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit([[0.0],[1.0]], {'Z':[[0.5],[0.5]]})
+ mm.fit([[0.0], [1.0]], {'Z': [[0.5], [0.5]]})
expdes = ExpDesigns(inp)
mod = PL()
mod.mc_reference['mean'] = [0.5]
@@ -100,8 +100,9 @@ def test__error_Mean_Std() -> None:
engine = Engine(mm, mod, expdes)
engine.start_engine()
mean, std = engine._error_Mean_Std()
- assert mean < 0.01 and std <0.01
-
+ assert mean < 0.01 and std < 0.01
+
+
#%% Test Engine._validError
def test__validError() -> None:
@@ -111,19 +112,20 @@ def test__validError() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit([[0.0],[1.0]], {'Z':[[0.5],[0.5]]})
+ mm.fit([[0.0], [1.0]], {'Z': [[0.5], [0.5]]})
expdes = ExpDesigns(inp)
mod = PL()
expdes.valid_samples = [[0.5]]
- expdes.valid_model_runs = {'Z':[[0.5]]}
+ expdes.valid_model_runs = {'Z': [[0.5]]}
mod.Output.names = ['Z']
engine = Engine(mm, mod, expdes)
engine.start_engine()
mean, std = engine._validError()
- assert mean['Z'][0] < 0.01 #and std['Z'][0] <0.01
-
+ assert mean['Z'][0] < 0.01 # and std['Z'][0] <0.01
+
+
#%% Test Engine._BME_Calculator
def test__BME_Calculator() -> None:
@@ -133,21 +135,22 @@ def test__BME_Calculator() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit([[0.0],[0.5],[1.0]], {'Z':[[0.5],[0.4],[0.5]]})
+ mm.fit([[0.0], [0.5], [1.0]], {'Z': [[0.5], [0.4], [0.5]]})
expdes = ExpDesigns(inp)
- expdes.generate_ED(2,transform=True,max_pce_deg=1)
+ expdes.generate_ED(2, max_pce_deg=1)
mod = PL()
mod.Output.names = ['Z']
engine = Engine(mm, mod, expdes)
engine.start_engine()
- obs_data = {'Z':np.array([0.45])}
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
+ obs_data = {'Z': np.array([0.45])}
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
engine._BME_Calculator(obs_data, sigma2Dict)
# Note: if error appears here it might also be due to inoptimal choice of training samples
+
def test__BME_Calculator_rmse() -> None:
"""
Calculate BME with given RMSE
@@ -155,21 +158,22 @@ def test__BME_Calculator_rmse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit([[0.0],[0.5],[1.0]], {'Z':[[0.5],[0.4],[0.5]]})
+ mm.fit([[0.0], [0.5], [1.0]], {'Z': [[0.5], [0.4], [0.5]]})
expdes = ExpDesigns(inp)
- expdes.generate_ED(2,transform=True,max_pce_deg=1)
+ expdes.generate_ED(2, max_pce_deg=1)
mod = PL()
mod.Output.names = ['Z']
engine = Engine(mm, mod, expdes)
engine.start_engine()
- obs_data = {'Z':np.array([0.45])}
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- engine._BME_Calculator(obs_data, sigma2Dict, rmse = {'Z':0.1})
+ obs_data = {'Z': np.array([0.45])}
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine._BME_Calculator(obs_data, sigma2Dict, rmse={'Z': 0.1})
# Note: if error appears here it might also be due to inoptimal choice of training samples
+
def test__BME_Calculator_lik() -> None:
"""
Calculate BME with given validation likelihood and post-snapshot
@@ -177,23 +181,23 @@ def test__BME_Calculator_lik() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit([[0.0],[0.5],[1.0]], {'Z':[[0.5],[0.4],[0.5]]})
+ mm.fit([[0.0], [0.5], [1.0]], {'Z': [[0.5], [0.4], [0.5]]})
expdes = ExpDesigns(inp)
- expdes.generate_ED(2,transform=True,max_pce_deg=1)
+ expdes.generate_ED(2, max_pce_deg=1)
mod = PL()
mod.Output.names = ['Z']
engine = Engine(mm, mod, expdes)
engine.start_engine()
- obs_data = {'Z':np.array([0.45])}
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
+ obs_data = {'Z': np.array([0.45])}
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
expdes.post_snapshot = True
-
+
engine.valid_likelihoods = [0.1]
engine._BME_Calculator(obs_data, sigma2Dict)
-
+
def test__BME_Calculator_2d() -> None:
"""
@@ -202,28 +206,28 @@ def test__BME_Calculator_2d() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
inp.add_marginals()
inp.Marginals[1].dist_type = 'normal'
- inp.Marginals[1].parameters = [0,1]
+ inp.Marginals[1].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit([[0.0,0.0],[0.5,0.1],[1.0,0.9]], {'Z':[[0.5],[0.4],[0.5]]})
+ mm.fit([[0.0, 0.0], [0.5, 0.1], [1.0, 0.9]], {'Z': [[0.5], [0.4], [0.5]]})
expdes = ExpDesigns(inp)
- expdes.generate_ED(2,transform=True,max_pce_deg=1)
+ expdes.generate_ED(2, max_pce_deg=1)
mod = PL()
mod.n_obs = 1
mod.Output.names = ['Z']
engine = Engine(mm, mod, expdes)
engine.start_engine()
- obs_data = {'Z':np.array([0.45])}
- m_observations = obs_data
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
+ obs_data = {'Z': np.array([0.45])}
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
expdes.post_snapshot = True
-
+
engine.valid_likelihoods = [0.1]
engine._BME_Calculator(obs_data, sigma2Dict)
-
+
+
#%% Test hellinger_distance
def test_hellinger_distance_isnan() -> None:
@@ -232,112 +236,118 @@ def test_hellinger_distance_isnan() -> None:
"""
P = [0]
Q = [1]
- math.isnan(hellinger_distance(P,Q))
-
+ math.isnan(hellinger_distance(P, Q))
+
+
def test_hellinger_distance_0() -> None:
"""
Calculate Hellinger distance-0
"""
- P = [0,1,2]
- Q = [1,0,2]
- assert hellinger_distance(P,Q) == 0.0
-
+ P = [0, 1, 2]
+ Q = [1, 0, 2]
+ assert hellinger_distance(P, Q) == 0.0
+
+
def test_hellinger_distance_1() -> None:
"""
Calculate Hellinger distance-1
"""
- P = [0,1,2]
- Q = [0,0,0]
- assert hellinger_distance(P,Q) == 1.0
-
+ P = [0, 1, 2]
+ Q = [0, 0, 0]
+ assert hellinger_distance(P, Q) == 1.0
+
+
#%% Test Engine._normpdf
-
+
def test__normpdf() -> None:
"""
Likelihoods based on gaussian dist
"""
-
+
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
mod = PL()
mod.Output.names = ['Z']
-
- y_hat_pce = {'Z':np.array([[0.12]])}
- std_pce = {'Z':np.array([[0.05]])}
- obs_data = {'Z':np.array([0.1])}
- sigma2Dict = {'Z':np.array([0.05])}
- total_sigma2s = pd.DataFrame(sigma2Dict, columns = ['Z'])
-
+
+ y_hat_pce = {'Z': np.array([[0.12]])}
+ std_pce = {'Z': np.array([[0.05]])}
+ obs_data = {'Z': np.array([0.1])}
+ sigma2Dict = {'Z': np.array([0.05])}
+ total_sigma2s = pd.DataFrame(sigma2Dict, columns=['Z'])
+
engine = Engine(mm, mod, expdes)
engine.start_engine()
engine._normpdf(y_hat_pce, std_pce, obs_data, total_sigma2s)
-
+
+
def test__normpdf_rmse() -> None:
"""
Likelihoods based on gaussian dist with rmse
"""
-
+
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
mod = PL()
mod.Output.names = ['Z']
-
- y_hat_pce = {'Z':np.array([[0.12]])}
- std_pce = {'Z':np.array([[0.05]])}
- obs_data = {'Z':np.array([0.1])}
- sigma2Dict = {'Z':np.array([0.05])}
- total_sigma2s = pd.DataFrame(sigma2Dict, columns = ['Z'])
-
+
+ y_hat_pce = {'Z': np.array([[0.12]])}
+ std_pce = {'Z': np.array([[0.05]])}
+ obs_data = {'Z': np.array([0.1])}
+ sigma2Dict = {'Z': np.array([0.05])}
+ total_sigma2s = pd.DataFrame(sigma2Dict, columns=['Z'])
+
engine = Engine(mm, mod, expdes)
engine.start_engine()
- engine._normpdf(y_hat_pce, std_pce, obs_data, total_sigma2s, rmse = {'Z':0.1})
-
-
+ engine._normpdf(y_hat_pce, std_pce, obs_data, total_sigma2s, rmse={'Z': 0.1})
+
+
#%% Test Engine._posteriorPlot
def test__posteriorPlot() -> None:
"""
Plot posterior
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
- expdes.generate_ED(2,transform=True,max_pce_deg=1)
+ expdes.generate_ED(2, max_pce_deg=1)
mod = PL()
- posterior = np.array([[0],[0.1],[0.2]])
+ posterior = np.array([[0], [0.1], [0.2]])
engine = Engine(mm, mod, expdes)
engine._posteriorPlot(posterior, ['i'], 'Z')
-
+
+
def test__posteriorPlot_2d() -> None:
"""
Plot posterior for 2 params
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
inp.add_marginals()
inp.Marginals[1].dist_type = 'normal'
- inp.Marginals[1].parameters = [0,1]
+ inp.Marginals[1].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
- expdes.generate_ED(2,transform=True,max_pce_deg=1)
+ expdes.generate_ED(2, max_pce_deg=1)
mod = PL()
- posterior = np.array([[0,0],[0.1,1.0],[0.2,0.5]])
+ posterior = np.array([[0, 0], [0.1, 1.0], [0.2, 0.5]])
engine = Engine(mm, mod, expdes)
engine._posteriorPlot(posterior, ['i', 'j'], 'Z')
-
+
+
#%% Test logpdf
def test_logpdf() -> None:
@@ -345,7 +355,8 @@ def test_logpdf() -> None:
Calculate log-pdf
"""
logpdf(np.array([0.1]), np.array([0.2]), np.array([0.1]))
-
+
+
#%% Test Engine._corr_factor_BME
# TODO: not used again here?
@@ -355,7 +366,7 @@ def test_subdomain() -> None:
"""
Create subdomains from bounds
"""
- subdomain([(0,1),(0,1)], 2)
+ subdomain([(0, 1), (0, 1)], 2)
#%% Test Engine.tradeoff_weights
@@ -363,89 +374,94 @@ def test_subdomain() -> None:
def test_tradeoff_weights_None() -> None:
"""
Tradeoff weights with no scheme
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
mod = PL()
engine = Engine(mm, mod, expdes)
- weights = engine.tradeoff_weights(None, [[0],[1]], {'Z':[[0.4],[0.5]]})
+ weights = engine.tradeoff_weights(None, [[0], [1]], {'Z': [[0.4], [0.5]]})
assert weights[0] == 0 and weights[1] == 1
-
+
+
def test_tradeoff_weights_equal() -> None:
"""
Tradeoff weights with 'equal' scheme
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
mod = PL()
engine = Engine(mm, mod, expdes)
- weights = engine.tradeoff_weights('equal', [[0],[1]], {'Z':[[0.4],[0.5]]})
+ weights = engine.tradeoff_weights('equal', [[0], [1]], {'Z': [[0.4], [0.5]]})
assert weights[0] == 0.5 and weights[1] == 0.5
-
+
+
def test_tradeoff_weights_epsdecr() -> None:
"""
Tradeoff weights with 'epsilon-decreasing' scheme
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 3
- expdes.X = np.array([[0],[1]])
+ expdes.X = np.array([[0], [1]])
mod = PL()
engine = Engine(mm, mod, expdes)
- weights = engine.tradeoff_weights('epsilon-decreasing', expdes.X, {'Z':[[0.4],[0.5]]})
+ weights = engine.tradeoff_weights('epsilon-decreasing', expdes.X, {'Z': [[0.4], [0.5]]})
assert weights[0] == 1.0 and weights[1] == 0.0
-
+
+
def test_tradeoff_weights_adaptive() -> None:
"""
Tradeoff weights with 'adaptive' scheme
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 3
- expdes.X = np.array([[0],[1]])
+ expdes.X = np.array([[0], [1]])
mod = PL()
engine = Engine(mm, mod, expdes)
- weights = engine.tradeoff_weights('adaptive', expdes.X, {'Z':[[0.4],[0.5]]})
+ weights = engine.tradeoff_weights('adaptive', expdes.X, {'Z': [[0.4], [0.5]]})
assert weights[0] == 0.5 and weights[1] == 0.5
-
+
+
def test_tradeoff_weights_adaptiveit1() -> None:
"""
Tradeoff weights with 'adaptive' scheme for later iteration (not the first)
- """
+ """
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
mod = PL()
engine = Engine(mm, mod, expdes)
- engine._y_hat_prev, _ = mm.eval_metamodel(samples=np.array([[0.1],[0.2],[0.6]]))
+ engine._y_hat_prev, _ = mm.eval_metamodel(samples=np.array([[0.1], [0.2], [0.6]]))
engine.tradeoff_weights('adaptive', expdes.X, expdes.Y)
-
+
+
#%% Test Engine.choose_next_sample
def test_choose_next_sample() -> None:
@@ -455,25 +471,26 @@ def test_choose_next_sample() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='random'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'random'
+ expdes.exploit_method = 'Space-filling'
+ expdes.util_func = 'Space-filling'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_da_spaceparallel() -> None:
"""
Chooses new sample using dual-annealing and space-filling, parallel=True
@@ -481,26 +498,27 @@ def test_choose_next_sample_da_spaceparallel() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='dual-annealing'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'dual-annealing'
+ expdes.exploit_method = 'Space-filling'
+ expdes.util_func = 'Space-filling'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
engine.parallel = True
x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_da_spacenoparallel() -> None:
"""
Chooses new sample using dual-annealing and space-filling, parallel = False
@@ -508,26 +526,27 @@ def test_choose_next_sample_da_spacenoparallel() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='dual-annealing'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'dual-annealing'
+ expdes.exploit_method = 'Space-filling'
+ expdes.util_func = 'Space-filling'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
engine.parallel = False
x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_loo_space() -> None:
"""
Chooses new sample using all LOO-CV and space-filling
@@ -535,25 +554,26 @@ def test_choose_next_sample_loo_space() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='LOO-CV'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'LOO-CV'
+ expdes.exploit_method = 'Space-filling'
+ expdes.util_func = 'Space-filling'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_vor_space() -> None:
"""
Chooses new sample using voronoi, space-filling
@@ -561,25 +581,26 @@ def test_choose_next_sample_vor_space() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='voronoi'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'voronoi'
+ expdes.exploit_method = 'Space-filling'
+ expdes.util_func = 'Space-filling'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_space() -> None:
"""
Chooses new sample using all latin-hypercube, space-filling
@@ -587,25 +608,26 @@ def test_choose_next_sample_latin_space() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'Space-filling'
+ expdes.util_func = 'Space-filling'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_alphD() -> None:
"""
Chooses new sample using all latin-hypercube, alphabetic (D)
@@ -613,25 +635,26 @@ def test_choose_next_sample_latin_alphD() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='alphabetic'
- expdes.util_func='D-Opt'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'alphabetic'
+ expdes.util_func = 'D-Opt'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- x, nan = engine.choose_next_sample(var = expdes.util_func)
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ x, nan = engine.choose_next_sample(var=expdes.util_func)
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_alphK() -> None:
"""
Chooses new sample using all latin-hypercube, alphabetic (K)
@@ -639,25 +662,26 @@ def test_choose_next_sample_latin_alphK() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='alphabetic'
- expdes.util_func='K-Opt'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'alphabetic'
+ expdes.util_func = 'K-Opt'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- x, nan = engine.choose_next_sample(var = expdes.util_func)
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ x, nan = engine.choose_next_sample(var=expdes.util_func)
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_alphA() -> None:
"""
Chooses new sample using all latin-hypercube, alphabetic (A)
@@ -665,25 +689,26 @@ def test_choose_next_sample_latin_alphA() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='alphabetic'
- expdes.util_func='A-Opt'
-
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'alphabetic'
+ expdes.util_func = 'A-Opt'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- x, nan = engine.choose_next_sample(var = expdes.util_func)
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ x, nan = engine.choose_next_sample(var=expdes.util_func)
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_VarALM() -> None:
"""
Chooses new sample using all latin-hypercube, VarDesign (ALM)
@@ -691,26 +716,27 @@ def test_choose_next_sample_latin_VarALM() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='VarOptDesign'
- expdes.util_func='ALM'
-
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'VarOptDesign'
+ expdes.util_func = 'ALM'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- x, nan = engine.choose_next_sample(var = expdes.util_func)
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ x, nan = engine.choose_next_sample(var=expdes.util_func)
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_VarEIGF() -> None:
"""
Chooses new sample using all latin-hypercube, VarDesign (EIGF)
@@ -718,25 +744,26 @@ def test_choose_next_sample_latin_VarEIGF() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='VarOptDesign'
- expdes.util_func='EIGF'
-
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'VarOptDesign'
+ expdes.util_func = 'EIGF'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- x, nan = engine.choose_next_sample(var = expdes.util_func)
- assert x.shape[0]==1 and x.shape[1] == 1
+ x, nan = engine.choose_next_sample(var=expdes.util_func)
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
def test_choose_next_sample_latin_VarLOO() -> None:
"""
@@ -745,26 +772,27 @@ def test_choose_next_sample_latin_VarLOO() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='VarOptDesign'
- expdes.util_func='LOOCV'
-
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'VarOptDesign'
+ expdes.util_func = 'LOOCV'
+
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- x, nan = engine.choose_next_sample(var = expdes.util_func)
- assert x.shape[0]==1 and x.shape[1] == 1
-
+ x, nan = engine.choose_next_sample(var=expdes.util_func)
+ assert x.shape[0] == 1 and x.shape[1] == 1
+
+
def test_choose_next_sample_latin_BODMI() -> None:
"""
Chooses new sample using all latin-hypercube, BayesOptDesign (MI)
@@ -772,28 +800,29 @@ def test_choose_next_sample_latin_BODMI() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesOptDesign'
- expdes.util_func='MI'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesOptDesign'
+ expdes.util_func = 'MI'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
-
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
+
+
def test_choose_next_sample_latin_BODALC() -> None:
"""
Chooses new sample using all latin-hypercube, BayesOptDesign (ALC)
@@ -801,28 +830,29 @@ def test_choose_next_sample_latin_BODALC() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesOptDesign'
- expdes.util_func='ALC'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesOptDesign'
+ expdes.util_func = 'ALC'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
-
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
+
+
def test_choose_next_sample_latin_BODDKL() -> None:
"""
Chooses new sample using all latin-hypercube, BayesOptDesign (DKL)
@@ -830,29 +860,29 @@ def test_choose_next_sample_latin_BODDKL() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesOptDesign'
- expdes.util_func='DKL'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesOptDesign'
+ expdes.util_func = 'DKL'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
+
-
def test_choose_next_sample_latin_BODDPP() -> None:
"""
Chooses new sample using all latin-hypercube, BayesOptDesign (DPP)
@@ -860,29 +890,29 @@ def test_choose_next_sample_latin_BODDPP() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesOptDesign'
- expdes.util_func='DPP'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesOptDesign'
+ expdes.util_func = 'DPP'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
+
-
def test_choose_next_sample_latin_BODAPP() -> None:
"""
Chooses new sample using all latin-hypercube, BayesOptDesign (APP)
@@ -890,59 +920,59 @@ def test_choose_next_sample_latin_BODAPP() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesOptDesign'
- expdes.util_func='APP'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesOptDesign'
+ expdes.util_func = 'APP'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
-
-def test_choose_next_sample_latin_BODMI() -> None:
+
+def test_choose_next_sample_latin_BODMI_() -> None:
"""
Chooses new sample using all latin-hypercube, BayesOptDesign (MI)
"""
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesOptDesign'
- expdes.util_func='MI'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesOptDesign'
+ expdes.util_func = 'MI'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
+
-
def test_choose_next_sample_latin_BADBME() -> None:
"""
Chooses new sample using all latin-hypercube, BayesActDesign (BME)
@@ -950,28 +980,29 @@ def test_choose_next_sample_latin_BADBME() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesActDesign'
- expdes.util_func='BME'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesActDesign'
+ expdes.util_func = 'BME'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
engine.n_obs = 1
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
+
def test_choose_next_sample_latin_BADDKL() -> None:
"""
@@ -980,28 +1011,28 @@ def test_choose_next_sample_latin_BADDKL() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesActDesign'
- expdes.util_func='DKL'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesActDesign'
+ expdes.util_func = 'DKL'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
engine.n_obs = 1
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
def test_choose_next_sample_latin_BADinfEntropy() -> None:
@@ -1011,52 +1042,25 @@ def test_choose_next_sample_latin_BADinfEntropy() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
expdes = ExpDesigns(inp)
expdes.n_init_samples = 2
expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
expdes.tradeoff_scheme = 'equal'
- expdes.explore_method='latin-hypercube'
- expdes.exploit_method='BayesActDesign'
- expdes.util_func='infEntropy'
+ expdes.explore_method = 'latin-hypercube'
+ expdes.exploit_method = 'BayesActDesign'
+ expdes.util_func = 'infEntropy'
mm = MetaModel(inp)
mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
mod = PL()
engine = Engine(mm, mod, expdes)
engine.out_names = ['Z']
- engine.observations = {'Z':np.array([0.45])}
- #engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
- sigma2Dict = {'Z':np.array([0.05])}
- sigma2Dict = pd.DataFrame(sigma2Dict, columns = ['Z'])
+ engine.observations = {'Z': np.array([0.45])}
+ # engine.choose_next_sample(sigma2=None, n_candidates=5, var='DKL')
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
engine.n_obs = 1
- x, nan = engine.choose_next_sample(sigma2=sigma2Dict, var = expdes.util_func)
-
-
-if __name__ == '__main__':
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- expdes = ExpDesigns(inp)
- expdes.n_init_samples = 2
- expdes.n_max_samples = 4
- expdes.X = np.array([[0],[1],[0.5]])
- expdes.Y = {'Z':[[0.4],[0.5],[0.45]]}
- expdes.explore_method='dual-annealing'
- expdes.exploit_method='Space-filling'
- expdes.util_func='Space-filling'
-
- mm = MetaModel(inp)
- mm.fit(expdes.X, expdes.Y)
- expdes.generate_ED(expdes.n_init_samples, transform=True, max_pce_deg=np.max(mm.pce_deg))
- mod = PL()
- engine = Engine(mm, mod, expdes)
- engine.out_names = ['Z']
- engine.parallel = True
- x, nan = engine.choose_next_sample()
- assert x.shape[0]==1 and x.shape[1] == 1
-
- None
\ No newline at end of file
+ engine.choose_next_sample(sigma2=sigma2Dict, var=expdes.util_func)
diff --git a/tests/test_ExpDesign.py b/tests/test_ExpDesign.py
index 42f87663c2d843c4fa3a23e047270673501dbd4c..68255b3380881a8182ecd5a3de7411842fcedebd 100644
--- a/tests/test_ExpDesign.py
+++ b/tests/test_ExpDesign.py
@@ -131,6 +131,47 @@ def test_random_sampler() -> None:
exp = ExpDesigns(inp)
exp.random_sampler(4)
+def test_random_sampler_largedatanoJDist() -> None:
+ """
+ Sample randomly, init_param_space implicitly, more samples wanted than given, no JDist available
+ """
+ x = np.random.uniform(0,1,1000)
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].input_data = x
+ exp = ExpDesigns(inp)
+ with pytest.raises(AttributeError) as excinfo:
+ exp.random_sampler(100000)
+ assert str(excinfo.value) == 'Sampling cannot proceed, build ExpDesign with max_deg != 0 to create JDist!'
+
+def test_random_sampler_largedataJDist0() -> None:
+ """
+ Sample randomly, init_param_space implicitly, more samples wanted than given,
+ JDist available, priors given via samples
+ """
+ x = np.random.uniform(0,1,1000)
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].input_data = x
+ exp = ExpDesigns(inp)
+ exp.init_param_space(max_deg = 1)
+ exp.random_sampler(100000)
+
+def test_random_sampler_largedataJDist1() -> None:
+ """
+ Sample randomly, init_param_space implicitly, more samples wanted than given,
+ JDist available, prior distributions given
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0,1]
+ exp = ExpDesigns(inp)
+ exp.init_param_space(max_deg = 1)
+ exp.random_sampler(100000)
+
+
+
def test_random_sampler_rawdata() -> None:
"""
Sample randomly, init_param_space implicitly, has 2d raw data
@@ -338,7 +379,7 @@ def test_read_from_file_wrongcomp():
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = ExpDesigns(inp, sampling_method = 'user')
- exp.hdf5_file = 'tests/ExpDesign_testfile.hdf5'
+ exp.hdf5_file = 'ExpDesign_testfile.hdf5'
with pytest.raises(KeyError) as excinfo:
exp.read_from_file(['Out'])
assert str(excinfo.value) == "'Unable to open object (component not found)'"
@@ -352,13 +393,5 @@ def test_read_from_file():
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = ExpDesigns(inp, sampling_method = 'user')
- exp.hdf5_file = 'tests/ExpDesign_testfile.hdf5'
+ exp.hdf5_file = 'ExpDesign_testfile.hdf5'
exp.read_from_file(['Z'])
-
-if __name__ == '__main__':
- x = np.random.uniform(0,1,1000)
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].input_data = x
- exp = ExpDesigns(inp, sampling_method = 'PCM')
- exp.generate_ED(4)
\ No newline at end of file
diff --git a/tests/test_Input.py b/tests/test_Input.py
index 84b9b239cb1646ccc2a946fb7a76831f17d38c1f..41c0e5ab84a5178b1934504bc63b5d5c03d19078 100644
--- a/tests/test_Input.py
+++ b/tests/test_Input.py
@@ -9,10 +9,10 @@ Class Input:
@author: Rebecca Kohlhaas
"""
import sys
-sys.path.append("src/")
-import pytest
-from bayesvalidrox.surrogate_models.inputs import Marginal, Input
+from bayesvalidrox.surrogate_models.inputs import Input
+
+sys.path.append("src/")
def test_addmarginals() -> None:
diff --git a/tests/test_InputSpace.py b/tests/test_InputSpace.py
index 1b5a28fa3eb4b1ad11c8a666a9e98e2b0dbaa8b9..ae31f8e90d051e39dd67c2e55f900a3eb0e11958 100644
--- a/tests/test_InputSpace.py
+++ b/tests/test_InputSpace.py
@@ -10,13 +10,16 @@ Class InputSpace:
"""
import sys
-sys.path.append("src/")
import pytest
import numpy as np
from bayesvalidrox.surrogate_models.inputs import Input
from bayesvalidrox.surrogate_models.input_space import InputSpace
+sys.path.append("src/")
+sys.path.append("../src/")
+
+
#%% Test ExpDesign.check_valid_input
def test_check_valid_input_hasmarg() -> None:
@@ -28,6 +31,7 @@ def test_check_valid_input_hasmarg() -> None:
InputSpace(inp)
assert str(excinfo.value) == 'Cannot build distributions if no marginals are given'
+
def test_check_valid_input_haspriors() -> None:
"""
Distribution not built if no distribution set for the marginals
@@ -36,45 +40,43 @@ def test_check_valid_input_haspriors() -> None:
inp.add_marginals()
with pytest.raises(AssertionError) as excinfo:
InputSpace(inp)
- assert str(excinfo.value) == 'Not all marginals were provided priors'
-
+ assert str(excinfo.value) == 'Not all marginals were provided priors'
+
+
def test_check_valid_input_priorsmatch() -> None:
"""
Distribution not built if dist types do not align
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
inp.add_marginals()
inp.Marginals[1].dist_type = 'normal'
- inp.Marginals[1].parameters = [0,1]
+ inp.Marginals[1].parameters = [0, 1]
with pytest.raises(AssertionError) as excinfo:
InputSpace(inp)
assert str(excinfo.value) == 'Distributions cannot be built as the priors have different types'
+
def test_check_valid_input_samples() -> None:
"""
Design built correctly - samples
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
inp.add_marginals()
- inp.Marginals[1].input_data = x+2
- try:
- InputSpace(inp)
- except AssertionError:
- pytest.fail("ExpDesign raised AssertionError unexpectedly!")
- # TODO: check for better options to assert that no error at all occurred
-
+ inp.Marginals[1].input_data = x + 2
+ InputSpace(inp)
+
def test_check_valid_input_both() -> None:
"""
Design no built - samples and dist type given
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
@@ -83,7 +85,8 @@ def test_check_valid_input_both() -> None:
InputSpace(inp)
assert str(excinfo.value) == 'Both samples and distribution type are given. Please choose only one.'
-#def test_check_valid_input_distnotok() -> None:
+
+# def test_check_valid_input_distnotok() -> None:
# """
# Design built incorrectly - dist types without parameters
# """
@@ -95,7 +98,7 @@ def test_check_valid_input_both() -> None:
# with pytest.raises(AssertionError) as excinfo:
# exp = ExpDesigns(inp)
# assert str(excinfo.value) == 'Some distributions do not have characteristic values'
-
+
def test_check_valid_input_distok() -> None:
"""
Design built correctly - dist types
@@ -103,16 +106,13 @@ def test_check_valid_input_distok() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
inp.add_marginals()
inp.Marginals[1].dist_type = 'normal'
- inp.Marginals[1].parameters = [0,1]
- try:
- InputSpace(inp)
- except AssertionError:
- pytest.fail("ExpDesign raised AssertionError unexpectedly!")
- # TODO: check for better options to assert that no error at all occurred
-
+ inp.Marginals[1].parameters = [0, 1]
+ InputSpace(inp)
+
+
def test_check_valid_input_noapc() -> None:
"""
Design built correctly - no apc
@@ -120,11 +120,12 @@ def test_check_valid_input_noapc() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
inp.add_marginals()
inp.Marginals[1].dist_type = 'normal'
- inp.Marginals[1].parameters = [0,1]
- InputSpace(inp, meta_Model_type = 'gpe')
+ inp.Marginals[1].parameters = [0, 1]
+ InputSpace(inp, meta_Model_type='gpe')
+
#%% Test ExpDesign.build_polytypes
def test_build_polytypes_normalerr() -> None:
@@ -140,6 +141,7 @@ def test_build_polytypes_normalerr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
def test_build_polytypes_normal() -> None:
"""
Build dist 'normal'
@@ -147,11 +149,11 @@ def test_build_polytypes_normal() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
exp = InputSpace(inp)
exp.build_polytypes(False)
-
-
+
+
def test_build_polytypes_uniferr() -> None:
"""
Build dist 'unif' - too few params
@@ -165,6 +167,7 @@ def test_build_polytypes_uniferr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
def test_build_polytypes_unif() -> None:
"""
Build dist 'unif'
@@ -172,10 +175,11 @@ def test_build_polytypes_unif() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'unif'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
exp = InputSpace(inp)
exp.build_polytypes(False)
-
+
+
def test_build_polytypes_gammaerr() -> None:
"""
Build dist 'gamma' - too few params
@@ -189,6 +193,8 @@ def test_build_polytypes_gammaerr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
+# noinspection SpellCheckingInspection
def test_build_polytypes_gamma() -> None:
"""
Build dist 'gamma'
@@ -196,12 +202,14 @@ def test_build_polytypes_gamma() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'gamma'
- inp.Marginals[0].parameters = [0,1,0]
+ inp.Marginals[0].parameters = [0, 1, 0]
exp = InputSpace(inp)
with pytest.raises(ValueError) as excinfo:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Parameter values are not valid, please set differently'
-
+
+
+# noinspection SpellCheckingInspection
def test_build_polytypes_betaerr() -> None:
"""
Build dist 'beta' - too few params
@@ -215,6 +223,7 @@ def test_build_polytypes_betaerr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
def test_build_polytypes_beta() -> None:
"""
Build dist 'beta'
@@ -222,11 +231,12 @@ def test_build_polytypes_beta() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'beta'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.build_polytypes(False)
-
-
+
+
+# noinspection SpellCheckingInspection
def test_build_polytypes_lognormerr() -> None:
"""
Build dist 'lognorm' - too few params
@@ -240,6 +250,7 @@ def test_build_polytypes_lognormerr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
def test_build_polytypes_lognorm() -> None:
"""
Build dist 'lognorm'
@@ -247,11 +258,11 @@ def test_build_polytypes_lognorm() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'lognorm'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.build_polytypes(False)
-
-
+
+
def test_build_polytypes_exponerr() -> None:
"""
Build dist 'expon' - too few params
@@ -265,6 +276,7 @@ def test_build_polytypes_exponerr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
def test_build_polytypes_expon() -> None:
"""
Build dist 'expon'
@@ -272,11 +284,11 @@ def test_build_polytypes_expon() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'expon'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.build_polytypes(False)
-
-
+
+
def test_build_polytypes_weibullerr() -> None:
"""
Build dist 'weibull' - too few params
@@ -290,6 +302,7 @@ def test_build_polytypes_weibullerr() -> None:
exp.build_polytypes(False)
assert str(excinfo.value) == 'Distribution has too few parameters!'
+
def test_build_polytypes_weibull() -> None:
"""
Build dist 'weibull'
@@ -297,50 +310,52 @@ def test_build_polytypes_weibull() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'weibull'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.build_polytypes(False)
-
+
def test_build_polytypes_arbitrary() -> None:
"""
Build poly 'arbitrary'
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.build_polytypes(False)
-
+
+
def test_build_polytypes_rosenblatt() -> None:
"""
Build dist with rosenblatt
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.build_polytypes(True)
-
+
+
def test_build_polytypes_samples() -> None:
"""
Build dist from samples
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.build_polytypes(False)
-
-
+
+
def test_build_polytypes_samples2d() -> None:
"""
Build dist from samples - samples too high dim
"""
- x = np.random.uniform(0,1,(2,1000))
+ x = np.random.uniform(0, 1, (2, 1000))
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
@@ -348,21 +363,22 @@ def test_build_polytypes_samples2d() -> None:
with pytest.raises(ValueError) as excinfo:
exp.build_polytypes(False)
assert str(excinfo.value) == 'The samples provided to the Marginals should be 1D only'
-
-
+
+
#%% Test ExpDesign.init_param_space
def test_init_param_space_nomaxdegsample() -> None:
"""
Init param space without max_deg for given samples
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.init_param_space()
+
def test_init_param_space_nomaxdegdist() -> None:
"""
Init param space without max_deg for given dist
@@ -370,21 +386,23 @@ def test_init_param_space_nomaxdegdist() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'expon'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.init_param_space()
-
+
+
def test_init_param_space_maxdeg() -> None:
"""
Init param space with max_deg for given samples
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
-
+
+
def test_init_param_space_maxdegdist() -> None:
"""
Init param space with max_deg for given dist (not uniform)
@@ -392,10 +410,11 @@ def test_init_param_space_maxdegdist() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'expon'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
-
+
+
def test_init_param_space_maxdegdistunif() -> None:
"""
Init param space with max_deg for given dist (uniform)
@@ -403,20 +422,19 @@ def test_init_param_space_maxdegdistunif() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'unif'
- inp.Marginals[0].parameters = [0.5,1,2,3]
+ inp.Marginals[0].parameters = [0.5, 1, 2, 3]
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
-
-
-
+
+
#%% Test ExpDesign.transform
-
+
def test_transform_noparamspace() -> None:
"""
Call transform without a built JDist
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(2,1000))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (2, 1000))
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
@@ -424,12 +442,13 @@ def test_transform_noparamspace() -> None:
with pytest.raises(AttributeError) as excinfo:
exp.transform(y)
assert str(excinfo.value) == 'Call function init_param_space first to create JDist'
-
+
+
def test_transform_dimerrlow() -> None:
"""
Call transform with too few dimensions
"""
- x = np.random.uniform(0,1,1000)
+ x = np.random.uniform(0, 1, 1000)
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
@@ -438,13 +457,14 @@ def test_transform_dimerrlow() -> None:
with pytest.raises(AttributeError) as excinfo:
exp.transform(x)
assert str(excinfo.value) == 'X should have two dimensions'
-
+
+
def test_transform_dimerrhigh() -> None:
"""
Call transform with too many dimensions
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(1,1,1000))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (1, 1, 1000))
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
@@ -453,13 +473,14 @@ def test_transform_dimerrhigh() -> None:
with pytest.raises(AttributeError) as excinfo:
exp.transform(y)
assert str(excinfo.value) == 'X should have two dimensions'
-
+
+
def test_transform_dimerr0() -> None:
"""
Call transform with wrong X.shape[0]
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(2,1000))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (2, 1000))
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
@@ -467,27 +488,30 @@ def test_transform_dimerr0() -> None:
exp.init_param_space(max_deg=2)
with pytest.raises(AttributeError) as excinfo:
exp.transform(y)
- assert str(excinfo.value) == 'The second dimension of X should be the same size as the number of marginals in the InputObj'
-
+ assert str(
+ excinfo.value) == 'The second dimension of X should be the same size as the number of marginals in the InputObj'
+
+
def test_transform_paramspace() -> None:
"""
Transform successfully
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(1000,1))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
exp.transform(y)
-
+
+
def test_transform_rosenblatt() -> None:
"""
Transform with rosenblatt
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(1000,1))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.Rosenblatt = True
inp.add_marginals()
@@ -495,86 +519,92 @@ def test_transform_rosenblatt() -> None:
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
exp.transform(y)
-
+
+
def test_transform_user() -> None:
"""
Transform with method 'user'
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(1000,1))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
- exp.transform(y, method = 'user')
-
+ exp.transform(y, method='user')
+
+
+# noinspection SpellCheckingInspection
def test_transform_rosenblattuser() -> None:
"""
Transform with rosenblatt and method 'user'
"""
- x = np.random.uniform(0,1,1000)
- y = np.random.uniform(0,1,(1000,1))
+ x = np.random.uniform(0, 1, 1000)
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.Rosenblatt = True
inp.add_marginals()
inp.Marginals[0].input_data = x
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
- exp.transform(y, method = 'user')
-
+ exp.transform(y, method='user')
+
+
def test_transform_uniform() -> None:
"""
Transform uniform dist
"""
- y = np.random.uniform(0,1,(1000,1))
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'unif'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
exp.transform(y)
-
+
+
def test_transform_norm() -> None:
"""
Transform normal dist
"""
- y = np.random.uniform(0,1,(1000,1))
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'norm'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
exp.transform(y)
-
+
+
# TODO: what are these other params here???
def test_transform_gammanoparam() -> None:
"""
Transform gamma dist - no parameters
"""
- y = np.random.uniform(0,1,(1000,1))
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'gamma'
- inp.Marginals[0].parameters = [1,1,0]
+ inp.Marginals[0].parameters = [1, 1, 0]
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
with pytest.raises(AttributeError) as excinfo:
exp.transform(y)
assert str(excinfo.value) == 'Additional parameters have to be set for the gamma distribution!'
-
+
+
def test_transform_gammaparam() -> None:
"""
Transform gamma dist - with parameters
"""
- y = np.random.uniform(0,1,(1000,1))
+ y = np.random.uniform(0, 1, (1000, 1))
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'gamma'
- inp.Marginals[0].parameters = [1,1,0]
+ inp.Marginals[0].parameters = [1, 1, 0]
exp = InputSpace(inp)
exp.init_param_space(max_deg=2)
- exp.transform(y, params = [1,1])
-
\ No newline at end of file
+ exp.transform(y, params=[1, 1])
diff --git a/tests/test_MCMC.py b/tests/test_MCMC.py
new file mode 100644
index 0000000000000000000000000000000000000000..3485a615b34150f63f8c8f428167453393531c09
--- /dev/null
+++ b/tests/test_MCMC.py
@@ -0,0 +1,223 @@
+# -*- coding: utf-8 -*-
+"""
+Test the MCM class of bayesvalidrox
+Tests are available for the following functions
+_check_ranges - x
+gelmain_rubin
+_iterative_scheme
+_my_ESS - x
+Class MCMC:
+ run_sampler
+ log_prior
+ log_likelihood
+ log_posterior
+ eval_model
+ train_error_model
+ marginal_llk_emcee
+"""
+import sys
+import pandas as pd
+import numpy as np
+
+from bayesvalidrox.surrogate_models.inputs import Input
+from bayesvalidrox.surrogate_models.exp_designs import ExpDesigns
+from bayesvalidrox.surrogate_models.surrogate_models import MetaModel
+from bayesvalidrox.pylink.pylink import PyLinkForwardModel as PL
+from bayesvalidrox.surrogate_models.engine import Engine
+from bayesvalidrox.bayes_inference.discrepancy import Discrepancy
+from bayesvalidrox.bayes_inference.mcmc import MCMC
+from bayesvalidrox.bayes_inference.bayes_inference import BayesInference
+from bayesvalidrox.bayes_inference.mcmc import _check_ranges, gelman_rubin
+
+sys.path.append("src/")
+sys.path.append("../src/")
+
+
+#%% Test MCMC init
+
+def test_MCMC() -> None:
+ """
+ Construct an MCMC object
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0])
+
+ mm = MetaModel(inp)
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ disc = Discrepancy('')
+ disc.type = 'Gaussian'
+ disc.parameters = (obsData * 0.15) ** 2
+ disc.opt_sigma = 'B'
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = disc
+ bi.inference_method = 'mcmc'
+ bi.setup_inference()
+ MCMC(bi)
+
+
+#%% Test run_sampler
+
+def test_run_sampler() -> None:
+ """
+ Run short MCMC
+
+ Returns
+ -------
+ None
+ DESCRIPTION.
+
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ expdes.x_values = np.array([0])
+
+ mm = MetaModel(inp)
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+ engine = Engine(mm, mod, expdes)
+
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ disc = Discrepancy('')
+ disc.type = 'Gaussian'
+ disc.parameters = (obsData * 0.15) ** 2
+ disc.opt_sigma = 'B'
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = disc
+ bi.inference_method = 'mcmc'
+ bi.setup_inference()
+ total_sigma2s = {'Z': np.array([0.15])}
+ mcmc = MCMC(bi)
+ mcmc.nburn = 10
+ mcmc.nsteps = 50
+ mcmc.run_sampler(mod.observations, total_sigma2s)
+
+
+#%% Test log_prior
+
+#%% Test log_likelihood
+
+#%% Test log_posterior
+
+#%% Test eval_model
+
+#%% Test train_error_model
+
+#%% Test gelmain_rubin
+
+def test_gelman_rubin() -> None:
+ """
+ Calculate gelman-rubin
+ """
+ chain = [[[1], [2]]]
+ gelman_rubin(chain)
+
+
+def test_gelman_rubin_returnvar() -> None:
+ """
+ Calculate gelman-rubin returning var
+ """
+ chain = [[[1], [2]]]
+ gelman_rubin(chain, return_var=True)
+
+
+#%% Test marginal_llk_emcee
+
+#%% Test _check_ranges
+
+def test_check_ranges() -> None:
+ """
+ Check to see if theta lies in expected ranges
+ """
+ theta = [0.5, 1.2]
+ ranges = [[0, 1], [1, 2]]
+ assert _check_ranges(theta, ranges) is True
+
+
+def test_check_ranges_inv() -> None:
+ """
+ Check to see if theta lies not in expected ranges
+ """
+ theta = [1.5, 1.2]
+ ranges = [[0, 1], [1, 2]]
+ assert _check_ranges(theta, ranges) is False
+
+
+#%% Main
+
+if __name__ == '__main__':
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+
+ expdes = ExpDesigns(inp)
+ expdes.n_init_samples = 2
+ expdes.n_max_samples = 4
+ expdes.X = np.array([[0], [1], [0.5]])
+ expdes.Y = {'Z': [[0.4], [0.5], [0.45]]}
+ # expdes.x_values = np.array([0]) # Error in plots if this is not available
+
+ mm = MetaModel(inp)
+ mm.fit(expdes.X, expdes.Y)
+ expdes.generate_ED(expdes.n_init_samples, max_pce_deg=np.max(mm.pce_deg))
+
+ mod = PL()
+ mod.observations = {'Z': np.array([0.45]), 'x_values': np.array([0])}
+ mod.Output.names = ['Z']
+
+ engine = Engine(mm, mod, expdes)
+
+ sigma2Dict = {'Z': np.array([0.05])}
+ sigma2Dict = pd.DataFrame(sigma2Dict, columns=['Z'])
+ obsData = pd.DataFrame(mod.observations, columns=mod.Output.names)
+ disc = Discrepancy('')
+ disc.type = 'Gaussian'
+ disc.parameters = (obsData * 0.15) ** 2
+ disc.opt_sigma = 'B'
+
+ bi = BayesInference(engine)
+ bi.Discrepancy = disc
+ bi.inference_method = 'mcmc'
+ bi.setup_inference()
+
+ # chain = [[[1],[2]]]
+ total_sigma2s = {'Z': np.array([0.15])}
+ mcmc = MCMC(bi)
+ mcmc.nsteps = 50
+ mcmc.nburn = 10
+ mcmc.run_sampler(mod.observations, total_sigma2s)
+ # mcmc.gelmain_rubin(chain)
+
+ chain = [[[1], [2]]]
+ gelman_rubin(chain, return_var=True)
diff --git a/tests/test_MetaModel.py b/tests/test_MetaModel.py
index b5095fa3494619b32af5d3ff81991e123adceab9..a3f9b19d3fb42c3e6ec3f34b3efb19298c8de545 100644
--- a/tests/test_MetaModel.py
+++ b/tests/test_MetaModel.py
@@ -26,13 +26,16 @@ Class MetaModel:
"""
import numpy as np
-import sys
-sys.path.append("src/")
import pytest
+import sys
from bayesvalidrox.surrogate_models.inputs import Input
from bayesvalidrox.surrogate_models.input_space import InputSpace
-from bayesvalidrox.surrogate_models.surrogate_models import MetaModel
+from bayesvalidrox.surrogate_models.surrogate_models import MetaModel, corr_loocv_error, create_psi
+from bayesvalidrox.surrogate_models.surrogate_models import gaussian_process_emulator
+
+sys.path.append("src/")
+
#%% Test MetaMod constructor on its own
@@ -43,9 +46,10 @@ def test_metamod() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
MetaModel(inp)
-
+
+
#%% Test MetaModel.build_metamodel
def test_build_metamodel_nosamples() -> None:
@@ -55,12 +59,12 @@ def test_build_metamodel_nosamples() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
with pytest.raises(AttributeError) as excinfo:
mm.build_metamodel()
assert str(excinfo.value) == 'Please provide samples to the metamodel before building it.'
-
+
def test_build_metamodel() -> None:
"""
@@ -69,11 +73,12 @@ def test_build_metamodel() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.CollocationPoints = [[0.2],[0.8]]
+ mm.CollocationPoints = np.array([[0.2], [0.8]])
mm.build_metamodel()
-
+
+
def test_build_metamodel_ninitsamples() -> None:
"""
Build MetaModel with n_init_samples
@@ -81,11 +86,12 @@ def test_build_metamodel_ninitsamples() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.CollocationPoints = [[0.2],[0.8]]
- mm.build_metamodel(n_init_samples = 2)
-
+ mm.CollocationPoints = np.array([[0.2], [0.8]])
+ mm.build_metamodel(n_init_samples=2)
+
+
def test_build_metamodel_gpe() -> None:
"""
Build MetaModel gpe
@@ -93,13 +99,13 @@ def test_build_metamodel_gpe() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.meta_model_type = 'gpe'
- mm.CollocationPoints = [[0.2],[0.8]]
+ mm.CollocationPoints = np.array([[0.2], [0.8]])
mm.build_metamodel()
-
-
+
+
def test_build_metamodel_coldimerr() -> None:
"""
Build MetaModel with wrong shape collocation samples
@@ -107,29 +113,31 @@ def test_build_metamodel_coldimerr() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.CollocationPoints = [[0.2,0.8]]
+ mm.CollocationPoints = [[0.2, 0.8]]
with pytest.raises(AttributeError) as excinfo:
mm.build_metamodel()
- assert str(excinfo.value) == 'The given samples do not match the given number of priors. The samples should be a 2D array of size (#samples, #priors)'
+ assert str(
+ excinfo.value) == 'The second dimension of X should be the same size as the number of marginals in the InputObj'
#%% Test MetaMod.generate_polynomials
-def test_generate_polynomials_noExp() -> None:
+def test_generate_polynomials_noexp() -> None:
"""
Generate polynomials without ExpDeg
"""
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
with pytest.raises(AttributeError) as excinfo:
mm.generate_polynomials()
assert str(excinfo.value) == 'Generate or add InputSpace before generating polynomials'
-
+
+
def test_generate_polynomials_nodeg() -> None:
"""
Generate polynomials without max_deg
@@ -137,21 +145,21 @@ def test_generate_polynomials_nodeg() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
-
+
# Setup
mm.InputSpace = InputSpace(inp)
mm.InputSpace.n_init_samples = 2
mm.InputSpace.init_param_space(np.max(mm.pce_deg))
mm.ndim = mm.InputSpace.ndim
mm.n_params = len(mm.input_obj.Marginals)
-
+
# Generate
with pytest.raises(AttributeError) as excinfo:
mm.generate_polynomials()
assert str(excinfo.value) == 'MetaModel cannot generate polynomials in the given scenario!'
-
+
def test_generate_polynomials_deg() -> None:
"""
@@ -160,96 +168,100 @@ def test_generate_polynomials_deg() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
-
+
# Setup
mm.InputSpace = InputSpace(inp)
mm.InputSpace.n_init_samples = 2
mm.InputSpace.init_param_space(np.max(mm.pce_deg))
mm.ndim = mm.InputSpace.ndim
mm.n_params = len(mm.input_obj.Marginals)
-
+
# Generate
mm.generate_polynomials(4)
-
-
+
+
#%% Test MetaMod.add_InputSpace
-def test_add_InputSpace() -> None:
+def test_add_inputspace() -> None:
"""
Add InputSpace in MetaModel
"""
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.add_InputSpace()
-
+
+
#%% Test MetaModel.fit
# Faster without these
-if 0:
- def test_fit() -> None:
- """
- Fit MetaModel
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]})
-
- def test_fit_parallel() -> None:
- """
- Fit MetaModel in parallel
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]}, parallel = True)
-
- def test_fit_verbose() -> None:
- """
- Fit MetaModel verbose
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]}, verbose = True)
-
-
- def test_fit_pca() -> None:
- """
- Fit MetaModel verbose and with pca
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.dim_red_method = 'pca'
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4,0.4],[0.5,0.6]]}, verbose = True)
-
- def test_fit_gpe() -> None:
- """
- Fit MetaModel
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.meta_model_type = 'gpe'
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]})
-
+def test_fit() -> None:
+ """
+ Fit MetaModel
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = MetaModel(inp)
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]})
+
+
+def test_fit_parallel() -> None:
+ """
+ Fit MetaModel in parallel
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = MetaModel(inp)
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]}, parallel=True)
+
+
+def test_fit_verbose() -> None:
+ """
+ Fit MetaModel verbose
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = MetaModel(inp)
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]}, verbose=True)
+
+
+def test_fit_pca() -> None:
+ """
+ Fit MetaModel verbose and with pca
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = MetaModel(inp)
+ mm.dim_red_method = 'pca'
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4, 0.4], [0.5, 0.6]]}, verbose=True)
+
+
+def test_fit_gpe() -> None:
+ """
+ Fit MetaModel
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = MetaModel(inp)
+ mm.meta_model_type = 'gpe'
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]})
+
+
#%% Test MetaModel.create_psi
-
+
def test_create_psi() -> None:
"""
Create psi-matrix
@@ -257,16 +269,16 @@ def test_create_psi() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.2],[0.8]])
+ samples = np.array([[0.2], [0.8]])
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- mm.create_psi(BasisIndices, univ_bas)
-
-
+ create_psi(BasisIndices, univ_bas)
+
+
#%% Test MetaModel.regression
def test_regression() -> None:
@@ -276,19 +288,20 @@ def test_regression() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
mm.regression(samples, outputs, psi)
-
+
+
def test_regression_ols() -> None:
"""
Regression: ols
@@ -296,19 +309,20 @@ def test_regression_ols() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'ols')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ols')
+
+
def test_regression_olssparse() -> None:
"""
Regression: ols and sparse
@@ -316,19 +330,20 @@ def test_regression_olssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'ols', sparsity = True)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ols', sparsity=True)
+
+
def test_regression_ard() -> None:
"""
Regression: ard
@@ -336,19 +351,20 @@ def test_regression_ard() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.2],[0.8]])
- outputs = np.array([0.4,0.5])
-
+ samples = np.array([[0.2], [0.8]])
+ outputs = np.array([0.4, 0.5])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'ard')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ard')
+
+
def test_regression_ardssparse() -> None:
"""
Regression: ard and sparse
@@ -356,19 +372,20 @@ def test_regression_ardssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.2],[0.8]])
- outputs = np.array([0.4,0.5])
-
+ samples = np.array([[0.2], [0.8]])
+ outputs = np.array([0.4, 0.5])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'ard', sparsity = True)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ard', sparsity=True)
+
+
def test_regression_fastard() -> None:
"""
Regression: fastard
@@ -376,19 +393,20 @@ def test_regression_fastard() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'fastard')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='fastard')
+
+
def test_regression_fastardssparse() -> None:
"""
Regression: fastard and sparse
@@ -396,19 +414,20 @@ def test_regression_fastardssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'fastard', sparsity = True)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='fastard', sparsity=True)
+
+
def test_regression_brr() -> None:
"""
Regression: brr
@@ -416,19 +435,20 @@ def test_regression_brr() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'brr')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='brr')
+
+
def test_regression_brrssparse() -> None:
"""
Regression: brr and sparse
@@ -436,19 +456,20 @@ def test_regression_brrssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'brr', sparsity = True)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='brr', sparsity=True)
+
+
def test_regression_bcs() -> None:
"""
Regression: bcs
@@ -456,19 +477,20 @@ def test_regression_bcs() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],[0.8],[0.9]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9])
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7], [0.8], [0.9]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9])
mm.pce_deg = 3
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(mm.pce_deg)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'bcs')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='bcs')
+
+
def test_regression_bcsssparse() -> None:
"""
Regression: bcs and sparse
@@ -476,20 +498,20 @@ def test_regression_bcsssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],[0.8],[0.9],[1.0]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.1])
-
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7], [0.8], [0.9], [1.0]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'bcs', sparsity = True)
-
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='bcs', sparsity=True)
+
+
def test_regression_lars() -> None:
"""
Regression: lars
@@ -497,19 +519,20 @@ def test_regression_lars() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],[0.8],[0.9],[1.0]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.1])
-
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7], [0.8], [0.9], [1.0]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'lars')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='lars')
+
+
def test_regression_larsssparse() -> None:
"""
Regression: lars and sparse
@@ -517,19 +540,20 @@ def test_regression_larsssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],[0.8],[0.9],[1.0]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.1])
-
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7], [0.8], [0.9], [1.0]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'lars', sparsity = True)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='lars', sparsity=True)
+
+
def test_regression_sgdr() -> None:
"""
Regression: sgdr
@@ -537,19 +561,20 @@ def test_regression_sgdr() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'sgdr')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='sgdr')
+
+
def test_regression_sgdrssparse() -> None:
"""
Regression: sgdr and sparse
@@ -557,20 +582,20 @@ def test_regression_sgdrssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'sgdr', sparsity = True)
-
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='sgdr', sparsity=True)
+
+
def test_regression_omp() -> None:
"""
Regression: omp
@@ -578,19 +603,20 @@ def test_regression_omp() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'omp')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='omp')
+
+
def test_regression_ompssparse() -> None:
"""
Regression: omp and sparse
@@ -598,20 +624,20 @@ def test_regression_ompssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'omp', sparsity = True)
-
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='omp', sparsity=True)
+
+
def test_regression_vbl() -> None:
"""
Regression: vbl
@@ -619,19 +645,20 @@ def test_regression_vbl() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'vbl')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='vbl')
+
+
def test_regression_vblssparse() -> None:
"""
Regression: vbl and sparse
@@ -639,19 +666,20 @@ def test_regression_vblssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'vbl', sparsity = True)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='vbl', sparsity=True)
+
+
def test_regression_ebl() -> None:
"""
Regression: ebl
@@ -659,19 +687,20 @@ def test_regression_ebl() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'ebl')
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ebl')
+
+
def test_regression_eblssparse() -> None:
"""
Regression: ebl and sparse
@@ -679,21 +708,20 @@ def test_regression_eblssparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.5])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method = 'ebl', sparsity = True)
-
-
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ebl', sparsity=True)
+
+
#%% Test Model.update_pce_coeffs
# TODO: very linked to the actual training...
@@ -707,15 +735,16 @@ def test_univ_basis_vals() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.2],[0.8]])
+ samples = np.array([[0.2], [0.8]])
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
mm.univ_basis_vals(samples)
-
+
+
#%% Test MetaModel.adaptive_regression
-
+
def test_adaptive_regression_fewsamples() -> None:
"""
Adaptive regression, no specific method, too few samples given
@@ -723,22 +752,25 @@ def test_adaptive_regression_fewsamples() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.8])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
# Evaluate the univariate polynomials on InputSpace
if mm.meta_model_type.lower() != 'gpe':
- mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
-
+ mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
+
with pytest.raises(AttributeError) as excinfo:
- mm.adaptive_regression(samples, outputs, 0)
- assert str(excinfo.value) == 'There are too few samples for the corrected loo-cv error. Fit surrogate on at least as many samples as parameters to use this'
-
+ mm.adaptive_regression(outputs, 0)
+ assert str(
+ excinfo.value) == ('There are too few samples for the corrected loo-cv error. Fit surrogate on at least as '
+ 'many samples as parameters to use this')
+
+
def test_adaptive_regression() -> None:
"""
Adaptive regression, no specific method
@@ -746,19 +778,20 @@ def test_adaptive_regression() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1]])
- outputs = np.array([0.0,0.1])
-
+ samples = np.array([[0.0], [0.1]])
+ outputs = np.array([0.0, 0.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
-
+ mm.build_metamodel(n_init_samples=2)
+
# Evaluate the univariate polynomials on InputSpace
if mm.meta_model_type.lower() != 'gpe':
- mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
- mm.adaptive_regression(samples, outputs, 0)
-
+ mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
+ mm.adaptive_regression(outputs, 0)
+
+
def test_adaptive_regression_verbose() -> None:
"""
Adaptive regression, no specific method, verbose output
@@ -766,19 +799,20 @@ def test_adaptive_regression_verbose() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1]])
- outputs = np.array([0.0,0.1])
-
+ samples = np.array([[0.0], [0.1]])
+ outputs = np.array([0.0, 0.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
-
+ mm.build_metamodel(n_init_samples=2)
+
# Evaluate the univariate polynomials on InputSpace
if mm.meta_model_type.lower() != 'gpe':
- mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
- mm.adaptive_regression(samples, outputs, 0, True)
-
+ mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
+ mm.adaptive_regression(outputs, 0, True)
+
+
def test_adaptive_regression_ols() -> None:
"""
Adaptive regression, ols
@@ -786,21 +820,22 @@ def test_adaptive_regression_ols() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],[0.8],
- [0.9],[1.0]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.1])
-
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7], [0.8],
+ [0.9], [1.0]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
-
+ mm.build_metamodel(n_init_samples=2)
+
# Evaluate the univariate polynomials on InputSpace
if mm.meta_model_type.lower() != 'gpe':
- mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
+ mm.univ_p_val = mm.univ_basis_vals(mm.CollocationPoints)
mm.pce_reg_method = 'ols'
- mm.adaptive_regression(samples, outputs, 0)
-
+ mm.adaptive_regression(outputs, 0)
+
+
#%% Test MetaModel.corr_loocv_error
def test_corr_loocv_error_nosparse() -> None:
@@ -810,22 +845,23 @@ def test_corr_loocv_error_nosparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],
- [0.8],[0.9],[1.0]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.1])
-
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7],
+ [0.8], [0.9], [1.0]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- outs = mm.regression(samples, outputs, psi, reg_method = 'ebl')
- mm.corr_loocv_error(outs['clf_poly'], outs['sparePsi'], outs['coeffs'],
- outputs)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ outs = mm.regression(samples, outputs, psi, reg_method='ebl')
+ corr_loocv_error(outs['clf_poly'], outs['sparePsi'], outs['coeffs'],
+ outputs)
+
+
def test_corr_loocv_error_singley() -> None:
"""
Corrected loocv error
@@ -833,21 +869,22 @@ def test_corr_loocv_error_singley() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
samples = np.array([[0.2]])
outputs = np.array([0.1])
-
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- outs = mm.regression(samples, outputs, psi, reg_method = 'ols')
- mm.corr_loocv_error(outs['clf_poly'], outs['sparePsi'], outs['coeffs'],
- outputs)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ outs = mm.regression(samples, outputs, psi, reg_method='ols')
+ corr_loocv_error(outs['clf_poly'], outs['sparePsi'], outs['coeffs'],
+ outputs)
+
+
def test_corr_loocv_error_sparse() -> None:
"""
Corrected loocv error from sparse results
@@ -855,25 +892,26 @@ def test_corr_loocv_error_sparse() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- samples = np.array([[0.0],[0.1],[0.2],[0.3],[0.4],[0.5],[0.6],[0.7],
- [0.8],[0.9],[1.0]])
- outputs = np.array([0.0,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.1])
-
+ samples = np.array([[0.0], [0.1], [0.2], [0.3], [0.4], [0.5], [0.6], [0.7],
+ [0.8], [0.9], [1.0]])
+ outputs = np.array([0.0, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.1])
+
mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
+ mm.build_metamodel(n_init_samples=2)
BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- outs = mm.regression(samples, outputs, psi, reg_method = 'ebl',
- sparsity = True)
- mm.corr_loocv_error(outs['clf_poly'], outs['sparePsi'], outs['coeffs'],
- outputs)
-
+ psi = create_psi(BasisIndices, univ_bas)
+
+ outs = mm.regression(samples, outputs, psi, reg_method='ebl',
+ sparsity=True)
+ corr_loocv_error(outs['clf_poly'], outs['sparePsi'], outs['coeffs'],
+ outputs)
+
+
#%% Test MetaModel.pca_transformation
-
+
def test_pca_transformation() -> None:
"""
Apply PCA
@@ -881,11 +919,12 @@ def test_pca_transformation() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- outputs = np.array([[0.4,0.4],[0.5,0.6]])
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
mm.pca_transformation(outputs)
+
def test_pca_transformation_verbose() -> None:
"""
Apply PCA verbose
@@ -893,11 +932,12 @@ def test_pca_transformation_verbose() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- outputs = np.array([[0.4,0.4],[0.5,0.6]])
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
mm.pca_transformation(outputs, True)
-
+
+
def test_pca_transformation_varcomp() -> None:
"""
Apply PCA with set var_pca_threshold
@@ -905,12 +945,13 @@ def test_pca_transformation_varcomp() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- outputs = np.array([[0.4,0.4],[0.5,0.6]])
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
mm.var_pca_threshold = 1
mm.pca_transformation(outputs)
-
+
+
def test_pca_transformation_ncomp() -> None:
"""
Apply PCA with set n_pca_components
@@ -918,9 +959,9 @@ def test_pca_transformation_ncomp() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- outputs = np.array([[0.4,0.4],[0.5,0.6]])
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
mm.n_pca_components = 1
mm.pca_transformation(outputs)
@@ -934,9 +975,9 @@ def test_gaussian_process_emulator() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.gaussian_process_emulator( [[0.2],[0.8]], [0.4,0.5])
+ inp.Marginals[0].parameters = [0, 1]
+ gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5])
+
def test_gaussian_process_emulator_nug() -> None:
"""
@@ -945,9 +986,9 @@ def test_gaussian_process_emulator_nug() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.gaussian_process_emulator( [[0.2],[0.8]], [0.4,0.5],nug_term=1.0)
+ inp.Marginals[0].parameters = [0, 1]
+ gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5], nug_term=1.0)
+
def test_gaussian_process_emulator_autosel() -> None:
"""
@@ -956,21 +997,21 @@ def test_gaussian_process_emulator_autosel() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.gaussian_process_emulator( [[0.2],[0.8]], [0.4,0.5],autoSelect=True)
+ inp.Marginals[0].parameters = [0, 1]
+ gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5], autoSelect=True)
-def test_gaussian_process_emulator_varIdx() -> None:
+
+def test_gaussian_process_emulator_varidx() -> None:
"""
Create GPE with var_idx
"""
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- mm.gaussian_process_emulator( [[0.2],[0.8]], [0.4,0.5],varIdx=1)
-
+ inp.Marginals[0].parameters = [0, 1]
+ gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5], varIdx=1)
+
+
#%% Test MetaModel.eval_metamodel
def test_eval_metamodel() -> None:
@@ -980,12 +1021,13 @@ def test_eval_metamodel() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.out_names = ['Z']
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]})
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]})
mm.eval_metamodel([[0.4]])
+
def test_eval_metamodel_normalboots() -> None:
"""
Eval trained MetaModel with normal bootstrap
@@ -993,13 +1035,14 @@ def test_eval_metamodel_normalboots() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.bootstrap_method = 'normal'
mm.out_names = ['Z']
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]})
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]})
mm.eval_metamodel([[0.4]])
+
def test_eval_metamodel_highnormalboots() -> None:
"""
Eval trained MetaModel with higher bootstrap-itrs
@@ -1007,13 +1050,14 @@ def test_eval_metamodel_highnormalboots() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.n_bootstrap_itrs = 2
mm.out_names = ['Z']
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]})
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]})
mm.eval_metamodel([[0.4]])
+
def test_eval_metamodel_gpe() -> None:
"""
Eval trained MetaModel - gpe
@@ -1021,14 +1065,14 @@ def test_eval_metamodel_gpe() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.meta_model_type = 'gpe'
mm.out_names = ['Z']
- mm.fit( [[0.2],[0.8]], {'Z':np.array([[0.4],[0.5]])})
+ mm.fit([[0.2], [0.8]], {'Z': np.array([[0.4], [0.5]])})
mm.eval_metamodel([[0.4]])
-
+
def test_eval_metamodel_pca() -> None:
"""
Eval trained MetaModel with pca
@@ -1036,13 +1080,14 @@ def test_eval_metamodel_pca() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.dim_red_method = 'pca'
mm.out_names = ['Z']
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4,0.4],[0.5,0.6]]})
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4, 0.4], [0.5, 0.6]]})
mm.eval_metamodel([[0.4]])
-
+
+
#%% Test MetaModel.create_model_error
# TODO: move model out of this function
@@ -1057,12 +1102,13 @@ def test_auto_vivification() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.auto_vivification()
-
+
+
#%% Test MetaModel.copy_meta_model_opts
-
+
def test_copy_meta_model_opts() -> None:
"""
Copy the metamodel with just some stats
@@ -1070,11 +1116,12 @@ def test_copy_meta_model_opts() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.add_InputSpace()
mm.copy_meta_model_opts()
-
+
+
#%% Test MetaModel.__select_degree
#%% Test Engine._compute_pce_moments
@@ -1086,11 +1133,12 @@ def test__compute_pce_moments() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4],[0.5]]})
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4], [0.5]]})
mm._compute_pce_moments()
+
def test__compute_pce_moments_pca() -> None:
"""
Compute moments of a pce-surrogate with pca
@@ -1098,12 +1146,13 @@ def test__compute_pce_moments_pca() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.dim_red_method = 'pca'
- mm.fit( [[0.2],[0.8]], {'Z':[[0.4,0.4],[0.5,0.6]]})
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4, 0.4], [0.5, 0.6]]})
mm._compute_pce_moments()
+
def test__compute_pce_moments_gpe() -> None:
"""
Compute moments of a gpe-surrogate
@@ -1111,30 +1160,13 @@ def test__compute_pce_moments_gpe() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
+ inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.meta_model_type = 'gpe'
with pytest.raises(AttributeError) as excinfo:
mm._compute_pce_moments()
assert str(excinfo.value) == 'Moments can only be computed for pce-type surrogates'
-
-#%% Test MetaModel.update_metamodel
-#TODO: taken from engine
-if __name__ == '__main__':
-
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0,1]
- mm = MetaModel(inp)
- samples = np.array([[0.2]])
- outputs = np.array([0.5])
-
- mm.CollocationPoints = samples
- mm.build_metamodel(n_init_samples = 2)
- BasisIndices = mm.allBasisIndices[str(1)][str(1.0)]
- univ_bas = mm.univ_basis_vals(samples)
- psi = mm.create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi)
+
+#%% Test MetaModel.update_metamodel
+# TODO: taken from engine