From 3d30ea005bf848098b626f176b40884e0b8a0f20 Mon Sep 17 00:00:00 2001
From: kohlhaasrebecca <rebecca.kohlhaas@outlook.com>
Date: Fri, 2 Aug 2024 11:24:41 +0200
Subject: [PATCH] Moved preprocessing for the training to a decorator
_decorator_fit
Preprocessing includes building the InputSpace, checking the input dimensions and assigning the inputs to parameters.
---
.../.$training restructure.drawio.bkp | 64 +++++++
docs/diagrams/training restructure.drawio | 70 +++++++
.../example_analytical_function.py | 16 +-
.../surrogate_models/polynomial_chaos.py | 107 ++++-------
.../surrogate_models/surrogate_models.py | 180 +++++++++++-------
5 files changed, 285 insertions(+), 152 deletions(-)
create mode 100644 docs/diagrams/.$training restructure.drawio.bkp
create mode 100644 docs/diagrams/training restructure.drawio
diff --git a/docs/diagrams/.$training restructure.drawio.bkp b/docs/diagrams/.$training restructure.drawio.bkp
new file mode 100644
index 000000000..bd3e47de1
--- /dev/null
+++ b/docs/diagrams/.$training restructure.drawio.bkp
@@ -0,0 +1,64 @@
+<mxfile host="Electron" modified="2024-07-31T10:12:06.359Z" 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="_svUbBalPQuec83DpYRD" version="22.1.11" type="device">
+ <diagram name="Page-1" id="63dxfbV-El8Ic4h_2P3G">
+ <mxGraphModel dx="1434" dy="956" 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="Kjr_m6IkcFhD3S5Vm94S-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="Kjr_m6IkcFhD3S5Vm94S-1" target="Kjr_m6IkcFhD3S5Vm94S-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-1" value="adaptive_regression" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="480" y="480" width="130" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-2" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;" vertex="1" parent="1">
+ <mxGeometry x="460" y="450" width="50" height="100" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-3" value="parallel" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="520" y="440" width="60" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-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="Kjr_m6IkcFhD3S5Vm94S-4" target="Kjr_m6IkcFhD3S5Vm94S-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-4" value="update" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="460" y="590" width="120" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-5" value="regression" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="670" y="540" width="120" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-9" value="b_i number, fast_bootstrap" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="490" y="410" width="180" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-10" value="storage" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="425" y="660" width="80" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-11" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;" vertex="1" parent="1">
+ <mxGeometry x="440" y="420" width="50" height="230" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-12" value="trafo_output[key]" style="rounded=0;whiteSpace=wrap;html=1;" vertex="1" parent="1">
+ <mxGeometry x="440" y="350" width="130" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-13" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;" vertex="1" parent="1">
+ <mxGeometry x="425" y="330" width="50" height="370" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-14" value="Outputs, keys" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="470" y="320" width="110" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-15" value="setup PCA" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="420" y="280" width="80" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-16" value="bootstrap step" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="430" y="250" width="80" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-18" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;" vertex="1" parent="1">
+ <mxGeometry x="410" y="240" width="50" height="480" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-19" value="bootstrap" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="450" y="220" width="110" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-20" value="setup bootstrap" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="400" y="190" width="110" height="30" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+</mxfile>
diff --git a/docs/diagrams/training restructure.drawio b/docs/diagrams/training restructure.drawio
new file mode 100644
index 000000000..c8e50029b
--- /dev/null
+++ b/docs/diagrams/training restructure.drawio
@@ -0,0 +1,70 @@
+<mxfile host="Electron" modified="2024-07-31T11:03:38.920Z" 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="tU7ui0JPg_6b6hPSW8yx" version="22.1.11" type="device">
+ <diagram name="Page-1" id="63dxfbV-El8Ic4h_2P3G">
+ <mxGraphModel dx="1434" dy="956" 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="Kjr_m6IkcFhD3S5Vm94S-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="Kjr_m6IkcFhD3S5Vm94S-1" target="Kjr_m6IkcFhD3S5Vm94S-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-1" value="adaptive_regression" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f8cecc;strokeColor=#b85450;" vertex="1" parent="1">
+ <mxGeometry x="480" y="480" width="130" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-2" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;" vertex="1" parent="1">
+ <mxGeometry x="460" y="450" width="50" height="100" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-3" value="parallel" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="520" y="440" width="60" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-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="Kjr_m6IkcFhD3S5Vm94S-4" target="Kjr_m6IkcFhD3S5Vm94S-5">
+ <mxGeometry relative="1" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-4" value="update" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f8cecc;strokeColor=#b85450;" vertex="1" parent="1">
+ <mxGeometry x="460" y="590" width="120" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-5" value="regression" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#f8cecc;strokeColor=#b85450;" vertex="1" parent="1">
+ <mxGeometry x="670" y="540" width="120" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-9" value="b_i number, fast_bootstrap" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#66B2FF;" vertex="1" parent="1">
+ <mxGeometry x="490" y="410" width="180" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-10" value="storage" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#FF3333;" vertex="1" parent="1">
+ <mxGeometry x="425" y="660" width="80" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-11" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;fillColor=#dae8fc;strokeColor=#6c8ebf;" vertex="1" parent="1">
+ <mxGeometry x="440" y="420" width="50" height="230" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-12" value="trafo_output[key]" style="rounded=0;whiteSpace=wrap;html=1;fillColor=#d5e8d4;strokeColor=#82b366;" vertex="1" parent="1">
+ <mxGeometry x="440" y="350" width="130" height="40" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-13" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;" vertex="1" parent="1">
+ <mxGeometry x="425" y="330" width="50" height="370" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-14" value="Outputs, keys" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="470" y="320" width="110" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-15" value="setup PCA" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#009900;" vertex="1" parent="1">
+ <mxGeometry x="420" y="280" width="80" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-16" value="bootstrap step" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#66B2FF;" vertex="1" parent="1">
+ <mxGeometry x="430" y="250" width="80" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-18" value="" style="strokeWidth=2;html=1;shape=mxgraph.flowchart.annotation_1;align=left;pointerEvents=1;fillColor=#dae8fc;strokeColor=#6c8ebf;" vertex="1" parent="1">
+ <mxGeometry x="410" y="240" width="50" height="480" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-19" value="bootstrap" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#66B2FF;" vertex="1" parent="1">
+ <mxGeometry x="450" y="230" width="110" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-20" value="setup bootstrap" style="text;html=1;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;fontColor=#66B2FF;" vertex="1" parent="1">
+ <mxGeometry x="400" y="200" width="110" height="30" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-21" value="" style="shape=curlyBracket;whiteSpace=wrap;html=1;rounded=1;labelPosition=left;verticalLabelPosition=middle;align=right;verticalAlign=middle;" vertex="1" parent="1">
+ <mxGeometry x="350" y="170" width="20" height="600" as="geometry" />
+ </mxCell>
+ <mxCell id="Kjr_m6IkcFhD3S5Vm94S-22" value="decorate_fit" style="text;html=1;strokeColor=none;fillColor=none;align=center;verticalAlign=middle;whiteSpace=wrap;rounded=0;" vertex="1" parent="1">
+ <mxGeometry x="365" y="140" width="110" height="30" as="geometry" />
+ </mxCell>
+ </root>
+ </mxGraphModel>
+ </diagram>
+</mxfile>
diff --git a/examples/analytical-function/example_analytical_function.py b/examples/analytical-function/example_analytical_function.py
index ceed4a5d6..4993f3383 100644
--- a/examples/analytical-function/example_analytical_function.py
+++ b/examples/analytical-function/example_analytical_function.py
@@ -173,22 +173,22 @@ if __name__ == "__main__":
# 1) MI (Mutual information) 2) ALC (Active learning McKay)
# 2)DKL (Kullback-Leibler Divergence) 3)DPP (D-Posterior-percision)
# 4)APP (A-Posterior-percision) # ['DKL', 'BME', 'infEntropy']
- # MetaModelOpts.ExpDesign.util_func = 'DKL'
+ # ExpDesign.util_func = 'DKL'
# BayesActDesign -> when data is available
# 1) BME (Bayesian model evidence) 2) infEntropy (Information entropy)
# 2)DKL (Kullback-Leibler Divergence)
- #MetaModelOpts.ExpDesign.util_func = 'DKL'
+ # ExpDesign.util_func = 'DKL'
# VarBasedOptDesign -> when data is not available
# 1)ALM 2)EIGF, 3)LOOCV
# or a combination as a list
- # MetaModelOpts.ExpDesign.util_func = 'EIGF'
+ # ExpDesign.util_func = 'EIGF'
# alphabetic
# 1)D-Opt (D-Optimality) 2)A-Opt (A-Optimality)
# 3)K-Opt (K-Optimality) or a combination as a list
- # MetaModelOpts.ExpDesign.util_func = 'D-Opt'
+ # ExpDesign.util_func = 'D-Opt'
# Defining the measurement error, if it's known a priori
obsData = pd.DataFrame(Model.observations, columns=Model.Output.names)
@@ -210,14 +210,8 @@ if __name__ == "__main__":
ExpDesign.valid_model_runs = {'Z': prior_outputs[:500]}
- # Run using the new engine
- #engine = Engine(MetaModelOpts, Model, ExpDesign)
- #engine.start_engine()
- #engine.train_normal()
-
- MetaModelOpts.ExpDesign = ExpDesign
+ # Run using the engine
engine = Engine(MetaModelOpts, Model, ExpDesign)
- engine.start_engine()
#engine.train_sequential()
engine.train_normal()
diff --git a/src/bayesvalidrox/surrogate_models/polynomial_chaos.py b/src/bayesvalidrox/surrogate_models/polynomial_chaos.py
index 0f921e052..4525837fb 100644
--- a/src/bayesvalidrox/surrogate_models/polynomial_chaos.py
+++ b/src/bayesvalidrox/surrogate_models/polynomial_chaos.py
@@ -8,28 +8,29 @@ import copy
import os
import warnings
+#from functools import wraps
import matplotlib.pyplot as plt
import numpy as np
-import scipy as sp
-import sklearn.gaussian_process.kernels as kernels
+#import scipy as sp
+#import sklearn.gaussian_process.kernels as kernels
import sklearn.linear_model as lm
from joblib import Parallel, delayed
-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 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 .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 .input_space import InputSpace
from .orthogonal_matching_pursuit import OrthogonalMatchingPursuit
from .reg_fast_ard import RegressionFastARD
from .reg_fast_laplace import RegressionFastLaplace
-from .surrogate_models import MetaModel
+from .surrogate_models import MetaModel, _decorator_fit
from .supplementary import corr_loocv_error, create_psi
@@ -60,7 +61,6 @@ class PCE(MetaModel):
_pce_reg_method : str
PCE regression method to compute the coefficients. The following
regression methods are available:
-
1. OLS: Ordinary Least Square method
2. BRR: Bayesian Ridge Regression
3. LARS: Least angle regression
@@ -105,12 +105,11 @@ class PCE(MetaModel):
# Use parent init
super().__init__(input_obj, bootstrap_method,
- n_bootstrap_itrs, dim_red_method, is_gaussian,
- verbose)
+ n_bootstrap_itrs, dim_red_method, is_gaussian, verbose)
# Additional inputs
# Parameters that are not needed from the outside are denoted with '_'
- self._meta_model_type = meta_model_type
+ self.meta_model_type = meta_model_type
self._pce_reg_method = pce_reg_method
self._pce_deg = pce_deg
self._pce_q_norm = pce_q_norm
@@ -133,6 +132,7 @@ class PCE(MetaModel):
self._q_norm_dict = None
self._deg_dict = None
+
def check_is_gaussian(self, _pce_reg_method, n_bootstrap_itrs) -> bool:
"""
Check if the metamodel returns a mean and stdev.
@@ -152,11 +152,13 @@ class PCE(MetaModel):
return False
else:
return True
+
-
- def build_metamodel(self, n_init_samples=None) -> None:
+ def build_metamodel(self) -> None:
"""
- Builds the parts for the metamodel (polynomes,...) that are neede before fitting.
+ Builds the parts for the metamodel (polynomes,...) that are needed before fitting.
+ This is executed outside of any loops related to e.g. bootstrap or
+ transformations such as pca.
Returns
-------
@@ -164,20 +166,6 @@ class PCE(MetaModel):
DESCRIPTION.
"""
-
- # Generate general warnings
- if 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)
-
- # TODO: does this work?
- super().build_metamodel()
-
# Generate polynomials
self._generate_polynomials(np.max(self._pce_deg))
@@ -189,18 +177,8 @@ class PCE(MetaModel):
self.LOOCV_score_dict = self.auto_vivification()
self._clf_poly = self.auto_vivification()
self._LCerror = self.auto_vivification()
- if self.dim_red_method.lower() == 'pca':
- self.pca = self.auto_vivification()
-
- # Define an array containing the degrees
- 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)')
- self._deg_array = self.__select_degree(ndim, self.n_samples)
+ self._deg_array = self.__select_degree(self.ndim, self.n_samples)
# Generate all basis indices
self._allBasisIndices = self.auto_vivification()
@@ -215,6 +193,11 @@ class PCE(MetaModel):
reverse=False, graded=True)
self._allBasisIndices[str(deg)][str(q)] = basis_indices
+ # Evaluate the univariate polynomials on InputSpace
+ self._univ_p_val = self.univ_basis_vals(self.CollocationPoints)
+
+ @_decorator_fit
+ #@_decorator_bootstrap
def fit(self, X: np.array, y: dict, parallel=False, verbose=False):
"""
Fits the surrogate to the given data (samples X, outputs y).
@@ -239,25 +222,9 @@ class PCE(MetaModel):
None.
"""
- # Use settings
- self.verbose = verbose
- self.parallel = parallel
-
- # TODO: other option: rebuild every time
- self.CollocationPoints = X
- if self._deg_array is None:
- self.build_metamodel(n_init_samples=X.shape[1])
-
- # Evaluate the univariate polynomials on InputSpace
- self._univ_p_val = self.univ_basis_vals(self.CollocationPoints)
- # Store the original ones for later use
+ # Store the original polynomial evaluation for later use
orig__univ_p_val = copy.deepcopy(self._univ_p_val)
- # --- Loop through data points and fit the surrogate ---
- if verbose:
- print(f"\n>>>> Training the {self._meta_model_type} metamodel "
- "started. <<<<<<\n")
-
# --- Bootstrap sampling ---
# Correct number of bootstrap if PCA transformation is required.
if self.dim_red_method.lower() == 'pca' and self.n_bootstrap_itrs == 1:
@@ -348,16 +315,13 @@ class PCE(MetaModel):
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.LOOCV_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._clf_poly[f'b_{b_i + 1}'][key][f"y_{i + 1}"] = out[i]['clf_poly']
# TODO: this is commented out here, but should be used in the SeqDesign??
# self._LCerror[f'b_{b_i+1}'][key][f"y_{i+1}"] = out[i]['_LCerror']
# Restore the univariate polynomials
self._univ_p_val = orig__univ_p_val
- 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):
@@ -387,9 +351,8 @@ class PCE(MetaModel):
# Make a copy
final_out_dict = copy.deepcopy(out_dict)
- # Create the _univ_p_val
- _univ_p_val = self._univ_p_val
-# _univ_p_val = self.univ_basis_vals(X, n_max=np.max(self._pce_deg))
+ # Create the _univ_p_val based on the given X (X expected to differ from self.CollocationPoints)
+ _univ_p_val = self.univ_basis_vals(X, n_max=np.max(self._pce_deg))
# Loop over the points
for i in range(y.shape[1]):
@@ -562,7 +525,7 @@ class PCE(MetaModel):
# Create a dict to pass the outputs
return_out_dict = dict()
- return_out_dict['_clf_poly'] = _clf_poly
+ return_out_dict['clf_poly'] = _clf_poly
return_out_dict['spareMulti-Index'] = sparse_basis_indices
return_out_dict['sparePsi'] = sparse_X
return_out_dict['coeffs'] = coeffs
@@ -633,27 +596,25 @@ class PCE(MetaModel):
# Assemble the Psi matrix
Psi = create_psi(BasisIndices, self._univ_p_val)
- #print(Psi.shape)
-
# Calulate the cofficients of the metamodel
outs = self.regression(Psi, ED_Y, BasisIndices)
# Calculate and save the score of LOOCV
- score, _LCerror = corr_loocv_error(outs['_clf_poly'],
+ 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']
- qAll_clf_poly[str(qidx + 1)] = outs['_clf_poly']
+ qAll_clf_poly[str(qidx + 1)] = outs['clf_poly']
qAllnTerms[str(qidx + 1)] = BasisIndices.shape[0]
qAll_LCerror[str(qidx + 1)] = _LCerror
@@ -755,7 +716,7 @@ class PCE(MetaModel):
# Create a dict to pass the outputs
returnVars = dict()
- returnVars['_clf_poly'] = _clf_poly
+ returnVars['clf_poly'] = _clf_poly
returnVars['degree'] = degree
returnVars['qnorm'] = qnorm
returnVars['coeffs'] = coeffs
@@ -796,7 +757,6 @@ class PCE(MetaModel):
samples,
n_max=np.max(self._pce_deg)
)
- #print(_univ_p_val)
mean_pred_b = {}
std_pred_b = {}
@@ -820,9 +780,6 @@ class PCE(MetaModel):
# Assemble Psi matrix
basis = self._basis_dict[f'b_{b_i + 1}'][output][in_key]
psi = create_psi(basis, _univ_p_val)
- #print(basis)
-
- #print(psi.shape)
# Prediction
if self.bootstrap_method != 'fast' or b_i == 0:
diff --git a/src/bayesvalidrox/surrogate_models/surrogate_models.py b/src/bayesvalidrox/surrogate_models/surrogate_models.py
index c80a6a4a4..a9a683876 100644
--- a/src/bayesvalidrox/surrogate_models/surrogate_models.py
+++ b/src/bayesvalidrox/surrogate_models/surrogate_models.py
@@ -8,29 +8,112 @@ import copy
import os
import warnings
+from functools import wraps
import matplotlib.pyplot as plt
import numpy as np
-import sklearn.linear_model as lm
-from joblib import Parallel, delayed
-from scipy.optimize import minimize, NonlinearConstraint
+#import sklearn.linear_model as lm
+#from joblib import Parallel, delayed
+#from scipy.optimize import minimize, NonlinearConstraint
from sklearn.decomposition import PCA as sklearnPCA
-from sklearn.preprocessing import MinMaxScaler
-from tqdm import tqdm
+#from sklearn.preprocessing import MinMaxScaler
+#from tqdm import tqdm
-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 .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 .reg_fast_ard import RegressionFastARD
+#from .reg_fast_laplace import RegressionFastLaplace
warnings.filterwarnings("ignore")
# Load the mplstyle
# noinspection SpellCheckingInspection
plt.style.use(os.path.join(os.path.split(__file__)[0],
'../', 'bayesvalidrox.mplstyle'))
+
+def _decorator_fit(fit_function):
+ """
+ This decorator performs the checks and preprocessing for fitting the
+ metamodel.
+
+ Parameters
+ ----------
+ fit_function : function
+ Function with the same signature as self.fit().
+
+ Raises
+ ------
+ AttributeError
+ DESCRIPTION.
+ ValueError
+ DESCRIPTION.
+
+ Returns
+ -------
+ decorator : function
+ The decorated function.
+
+ """
+ @wraps(fit_function)
+ def decorator(self , *args, **kwargs) :
+ # Use settings
+ # TODO: this a bit hacky, need to have arguments in exactly this order
+ X = args[0]
+ y = args[1]
+ self.verbose = args[2]
+ self.parallel = args[3]
+ self.ndim = len(self.input_obj.Marginals)
+
+ # Check X
+ X = np.array(X)
+ n_samples, ndim = X.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)')
+ self.CollocationPoints = X
+ self.n_samples = n_samples
+
+ # Check y
+ for key in y.keys():
+ y_val = np.array(y[key])
+ if y_val.ndim != 2:
+ raise ValueError('The given outputs y should be 2D')
+ y[key] = np.array(y[key])
+ self.out_names = list(y.keys())
+
+ # Build general parameters
+ if self.dim_red_method.lower() == 'pca':
+ self.pca = self.auto_vivification()
+
+ # Build the input space
+ if self.InputSpace is None:
+ self.InputSpace = InputSpace(self.input_obj)
+ n_init_samples = self.CollocationPoints.shape[0]
+ self.InputSpace.n_init_samples = n_init_samples
+ # TODO: _pce_deg not necessarily available, generalize this!
+ self.InputSpace.init_param_space(np.max(self._pce_deg))
+
+ # Transform input samples # TODO: Make 'method' variable
+ self.CollocationPoints = self.InputSpace.transform(self.CollocationPoints, method='user')
+
+ # Build any other surrogate parts
+ self.build_metamodel()
+
+ # --- Loop through data points and fit the surrogate ---
+ if self.verbose:
+ print(f"\n>>>> Training the {self._meta_model_type} metamodel "
+ "started. <<<<<<\n")
+
+ # Call the function for fitting
+ fit_function( self , *args, **kwargs)
+
+ if self.verbose:
+ print(f"\n>>>> Training the {self._meta_model_type} metamodel"
+ " sucessfully completed. <<<<<<\n")
+ return decorator
class MetaModel:
@@ -96,7 +179,13 @@ class MetaModel:
self.ndim = None
# Build general parameters
- self.InputSpace = InputSpace(self.input_obj)
+
+ # General warnings
+ if not self.is_gaussian:
+ print('There are no estimations of surrogate uncertainty available'
+ ' for the chosen regression options. This might lead to issues'
+ ' in later steps.')
+
def check_is_gaussian(self) -> bool:
"""
@@ -111,40 +200,9 @@ class MetaModel:
bool
True if stdev is also returned.
"""
- None
-
- def valid_training_inputs(self, X: np.array, y: dict):
- """
- Check if the training inputs are of the expected format.
- Will throw an error otherwise
-
- Parameters
- ----------
- X : 2D list or np.array of shape (#samples, #dim)
- 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.
-
- """
- # TODO: Add to this
- # Check size of inputs
- X = np.array(X)
- for key in y.keys():
- y_val = np.array(y[key])
- 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
+ pass
-
-
def build_metamodel(self) -> None:
"""
Builds the parts for the metamodel (polynomes,...) that are needed
@@ -158,24 +216,10 @@ class MetaModel:
None
"""
- # Extend input space
- n_init_samples=self.CollocationPoints.shape[1]
- if self.InputSpace.n_init_samples is None:
- if n_init_samples is None:
- n_init_samples = self._CollocationPoints.shape[0]
- self.InputSpace.n_init_samples = n_init_samples
- self.InputSpace.init_param_space(np.max(self._pce_deg))
- self.ndim = self.InputSpace.ndim
-
-
- # Transform input samples
- # TODO: Make 'method' variable
- self.CollocationPoints = self.InputSpace.transform(self.CollocationPoints, method='user')
-
- self.ndim = len(self.input_obj.Marginals)
-
-
-
+ pass
+
+
+ @_decorator_fit
def fit(self, X: np.array, y: dict, parallel=False, verbose=False):
"""
Fits the surrogate to the given data (samples X, outputs y).
@@ -200,10 +244,11 @@ class MetaModel:
None.
"""
- None
+ pass
+
# -------------------------------------------------------------------------
- def pca_transformation(self, target):
+ def pca_transformation(self, target, n_pca_components):
"""
Transforms the targets (outputs) via Principal Component Analysis.
The number of features is set by `self.n_pca_components`.
@@ -228,7 +273,7 @@ class MetaModel:
n_samples, n_features = target.shape
# Use the given n_pca_components
- n_pca_components = self.n_pca_components
+ #n_pca_components = self.n_pca_components
# Switch to var_pca if n_pca_components is too large
if (n_pca_components is not None) and (n_pca_components > n_features):
@@ -278,6 +323,7 @@ class MetaModel:
return pca, scaled_target, n_pca_components
+
# -------------------------------------------------------------------------
def eval_metamodel(self, samples):
"""
@@ -299,6 +345,7 @@ class MetaModel:
"""
return None, None
+
def compute_moments(self):
"""
Computes the first two moments of the metamodel.
@@ -347,6 +394,7 @@ class MetaModel:
new_MetaModelOpts.ndim = len(self.input_obj.Marginals)
return new_MetaModelOpts
+
# -------------------------------------------------------------------------
def add_InputSpace(self):
"""
--
GitLab