From 223c099cfc2051a2776938c3b29c1bb3e5cb48f1 Mon Sep 17 00:00:00 2001
From: kohlhaasrebecca <rebecca.kohlhaas@outlook.com>
Date: Fri, 23 Aug 2024 14:03:52 +0200
Subject: [PATCH] [fix] Changed InputSpace.apce to InputSpae.pce
---
.coverage | Bin 0 -> 53248 bytes
.../posterior/Z.npy | Bin 0 -> 176 bytes
src/bayesvalidrox/.coverage | Bin 0 -> 53248 bytes
.../surrogate_models/input_space.py | 15 +-
tests/test_InputSpace.py | 2 +-
tests/test_MetaModel.py | 927 +----------------
tests/test_PolynomialChaosEmulator.py | 952 ++++++++++++++++++
tests/test_supplementary.py | 162 +++
8 files changed, 1140 insertions(+), 918 deletions(-)
create mode 100644 .coverage
create mode 100644 Outputs_SeqPosteriorComparison/posterior/Z.npy
create mode 100644 src/bayesvalidrox/.coverage
create mode 100644 tests/test_PolynomialChaosEmulator.py
create mode 100644 tests/test_supplementary.py
diff --git a/.coverage b/.coverage
new file mode 100644
index 0000000000000000000000000000000000000000..2dac62dc3862970f94a717175f95a6889684b93f
GIT binary patch
literal 53248
zcmeI53ve98nSgtCX5WvNypq>?*cv_Tl@34S3&uD`!m=^O*s>ANSYf?djikl9JIl^4
zNMNiT+XwOil`E1Qq$*{btE(ebBwXt5s5rr3SDh|_s}x)w6-bfTiBc6fj$myifDX%=
z`@3gG+Fi*Muv#c`J<|3+)6@O`e}Da&meix}y*EbFlGrb&;^DNo9#tTYLmi@s5OTuD
z3m@~a!--&CfaLPdFL&xhtNY*b;#-l&_9()4db{u{&$qlQ+&}fioqy{-ZVx*naDyF;
z03*N%FarO-38;HL_CQSycld#HxHl%HQ{jk|(nD|OP1~-$d7F6iwOel7ChBowt4oAq
z)hh8?F(uzCCZ&|vAB{<3G|?B0gwxT)fS4YX%+wi0>LU{!#KBZY60WBC$yq4c2f5PH
z0H{c&qVaHQSUeyNucQeKYiVgHtw(?oDLRlKDa2MIdz+Y&`lXbVh)9a=MQgOL&DC~f
z#AOfo{oF$mEmJrZfp2#~X&_>@C_DN=QAAGl5$}6b;Y4ImQdWxLnG!~1@N+1wmo_EG
z*Q8}J7ERD%CX{G8Dknr~NQz|AQs2A<TuGpr2f)iO4d~iW3bbF(Y1qo6Gn;vyoJMjZ
zb3ZtjkWH+!8Hs7gOnpvd;=O?Uu1q{hDhG1thST!pa$YWPS~d#%k?Urjx|}U<L3NtQ
zNsa2cGKuKj8EK~SSBkSaSJGl6<dq^R1hjzVZd+iZpF_|pr6J|+7(8!<!kM&8uR~Dz
zp>+^=j#fJCfrbX|p}T2AA$UT)(X=vi;w)%1Gtrl6I^_Q4+YRX@g^h&tqF79n`)QXU
zxkbHlpf%_BnKdSiaEf#^Q8p5Sr=MwSBhG>kVsA|DEo^JF#V8VV5Eu*0Wc`^$gme{!
zlp!MDBc;Lv(rV+bJ}i9J4GEggI;SnLt>JP_XQl~+*3F!{)Eb*TP*=wtcG6~K)RrOa
z1x?7T$$5rR^Hy96E18rzkbKx6Y37(6CCxr!+=1e(vfBdJ)?KbRM!L{C<HVub1$&^n
znp5>2LB=KWu@)49#LRYqqKV+6yPpiF2gSL*lW1@}Xtf14RbS3=;zDR032jIAS*Xho
z)lm9z81%`9FL0T}%(}c&8eVlzI0mChG8|1omxVD{p8-UYurCaW!@Y7QP1`TIq|KAe
zJ+uDm3TCfO_K>QLCWr35M^=<5OsiztQm!^*#o%CAA)}BI6nkYkCWRCF+yFUtC}1<H
zh}4I<vr_x44rQ3=7`P`$eKIE(GR)r$ozCoEa1xJ)6Me>fMW$h0D~$c5a}qoHAOac1
zAaqIvx@eJ3Q>k&6j98}r)k$DNw8q1IdUq$4lBe<BL1~UUeKy(}PD|-%ToOqV>I`jI
z_l6ZQl9J#i%%yEX{==F9sjbjD5|UN2gFMQq4V*o2#TA^IF*>6@&V&@0)Z<~(ZR65X
z&`64I&bF5>7;xcfC%>H#c6m6+rEY-?Zh8&&>J-dW#<PyAjV$mvd=bGnI~V~*fDvE>
z7y(9r5nu!u0Y-okU<4QeM&K`x0MA)CJ6Zo*@b3`500BD~0Y-okU<4QeMt~7u1Q-EE
zfDvE>7y(A$^GU#C;g@3bPdwIQj;{+Ce*~~$_4?K8T~I0ueiPw0@w=bT0y3`|0Y-ok
zU<4QeMt~7u1Q-EEfDvE>7y(AWB;etfa^_C~EH1vzY5Wj?{QkcUJ%aFe@NK?}z9)T~
zysvr(y}ajf&z0_%-ID8F*CVcG=PBn6j$b=c4!ixNJ!pI0)@A)4>x0%>;VEGkKgEBS
z@8uU+UbGBb>bW0r_i<OC^Puu$AA8#bqy{^$?zu&QogY1^1M*;OFdSBTw#t!ATuP*s
zp6$`}^_kwD-te%bz@C?AUrHY8NlOacB4J3$B$F|Tq=g-vupQ)vR!~_}lFH;TY_2*0
zx^@LYS3_yK<k&FmZc@_3wG}11CM5=7pC0JCc{%7>Rg!Dn5^QKpq@&^3R!M=)t)O*Z
z3uxU`lGZ&?;Dj6pt(}8mS=koP#KLJgMQApcOmi1(wb>>2Nik4&eG{mwF3Hzjox3_g
z)9yyl)L4?H9f@Qn-JOKp%b;(E2>Jr0>m$q>N@ligC^=KVUCTgSb4i|UjViD`Dha!k
zNpHJ-DJcA6NeZ`+o=q#YQ;sLYsi-28rg(QfXdd|Znw3<fXt1C|M<|*w_G;^QLvhlK
zLa;q;*5qAvVDgTIFqxELWJ;!F*cq<CuI`?qYbai%7HsV-hpiA4HJT2Ez({Wm7}>E9
zMqo>5N*;hsaXL4pCrTQ;0ybhxz{Xd~Vk0FDg!;okG8Rt8^sY4+0Bg6D*IGE$2j#kV
zF<4T{Vo6S=2ju}d5sro66+DqaXd$6wCZ%MeWR{Ss!Ejeu3=hPlA-V@0%tWie%xz^c
zBf&f!f^D9mR0j59CWi<8Wvev3Z<D>yVKT+U7J;SmG{)r2ycUX(iH%sRESt6B{zfXS
zqHGmb{0udK!Bls7+Qv{aCa2&{9dwW%@RiLF9iT#rF(oIwU?^Oc5`{_2fQba2@lqN3
zh5@Lo0Z-WqCJiM+ee{8iw84Hi7}~oK1uLE=`wAxdkPGZ|mcvdw5|2PZ?s9^G1(|${
z3sQ6qwjvI&wIGvkk?x+n$v18XBcX*LRy-K=C^Jv)MoWM)_1eH(dFJ|=MMs|%Z0ssa
zk@CRZE`XI!@Ju=P$xmO%_2>c^eVGT%8yCWVLvx<j-4@V&#X{)LQ+g8zO4~lZ(!BNY
z7V>z#!+iJ!2;}$wZ3y1Yce_{cJRyAB{fcLU^&xAWZ_4)%zAw4nb12@|91eHCtHu7L
z>rw05_EyJlY%kdMc%#lAICtXj3cLAB{4;!n_gh}F+-F(J{g}I-TZw)PDnI^llb{B<
zlFC;+ceU*l)SA*%&a{`--2%MNp(OXn<Y2D<gQdGR(-l|TBB-lMb8VjW|MDI|-Bg;^
z%dh`iN~f8>{%_tPsMV$UYOMd8x&*bcG))ET|Hc~xHLw8sptFjlGs|E9H|!DA=F&VX
zT>me-Lr}k1nnF52&0YU5O$qA2C(t~17AT&h_3L@o|Mdq1^^OHGS+w4tXZ>H-C#XBi
zVk>X@sEr8fjs-C?=MiLP{a+Io)UTAs#yso)B~d}WrOej!^?zVUP?hpn`tbGt;(kHx
zDv#m(^?&tUf_htd%oMNxtIAhtdS9Md|NG0=80T64FDjq4;{FCxL1p<WtoRwKVo*@K
z%hWas*8jMCh6>mJK3P!1<tfp;>wj<g3RbxO_Y4T?-UTUG@ibYu{&(*e)XuWlG1mXC
z1)F?}3sQ6qUCFs%lW&pk{Pn*hA*i7RAyzyX%)9=#mv640S#;PWLETlJBIP}`SU=%2
z<=iK~;`P6<Pf#~5i2rlf|NPB@dc}h1&O>2=^>LfCqy^}f;`Kj!|KG>|F$K%P2rvSS
z03*N%FanGKBftnS0*nA7zzBSf2=E+_yyW-)9G*h(%??I@5nu!u0Y-okU<4QeMt~7u
z1Q-EEfD!l$B*61tC*A*VeIDTqj^RGv`@Rc4$@_Kh!=CSZwtN2C{g(UZ?!R~Ux!1a0
zbwyltPR;q>@OS^i&a0iRPSJ7HvBuHj2-wfte_>bcz4nc^AKKb&KI^pgSD=y|i~u9R
z2rvTWB2e2VaLy~+gx`Kpp><5vbxe81X)nqh>pD2jH@CvQ>*4h$6W1L*b*goB)15s%
zJ*Q8dy6W!n@!qW!jZG_>P}UcOM9YH0!R9+xE^cnBTC)PK_rr@xQ1+kY(UDOtJINuG
z<3V&xA@LCEn6jga=?8*S7k#;ram_xw+{!uC6uf=}oqe?-r;TZTZDf4p#AG`f)9O%8
zvt=ho_>QSFT!khhaxPAvN2A&XG_HBGlbWq#3T^|RjwubSXm)bK!3hs}?T`qAo(4qz
zq3+BvcFezB*eX2GLQLNW?<hgf;PK8W|H$}PCmY63OxAZC^ofT?PfQ*=F?nRNu405A
zmjTV_87{X$oVJaf&q5^8Mx=3*{<q$|L@0;HWuzz7Y~u7-`)K=kL+-1ScF^2j(LRoY
z&46kTyu<`0H;>|x(S|Xc^S9@~3q0Nh&WyIBT<4TJ$t`VygkOT!lbqHb96AFy;hP@4
zz2feA%icy(jHX5*SvTfCF?nLruYqI1=}K)3)U<=i?I567$4173(@-SxJ))PcaoVpP
zgD<~E0F8&H{Mx8?Y{V&o#VS$w_EG-I1~^T?YeUXe1|Bl(iql6X|K`lZ_No5>U%^Kx
z*P7`7a%3m1Xgtt9c4l(Tw0jxkTe3_@Y#n>p?p#Xr!Mi=yRCd=BZ&%b$w690m-@l?&
z5}hJ^C{U14IC5zoON^5aD#6RF5s{b7qSn-b*27<@o6vrQDojm!LJt+QxoAbr=$LcT
zF(+o;h(a5APQlBN>3?n}Jjs!@)k4{QwSyHkAe(9&JVGbk6+D{YH!p#hgG;W07;=E4
z4*j1#Tl9zz53|<<Ap4<!W5i5uqN0i*vl*YKhs)9n!a!KP81mn_*wIA@!CN=rFdG&*
zCx;`iU=>x>kp8c#9sT6a*)?atV_R0Jt%BGqt3cxuB%T~u5`Zelb$(LU9{-DLI+0)@
zQTxd)$PMN#1=WlU`)lca<ePQHfyC4zqNi`s=z#~%-9pVHG|@MOv~E;unP`8!|Drn!
z`Fwg_vlyh19cEO~d0sU5YgB*^AMzwY#ch?u=IYAPO|C5~z;QdxB67%Po=|B)whjxa
zI<Q6VWdHQW%Nrj-d(a?UK$#pCsD<pmI?zd!y*7&u3XU&S0I>dw7lR1Jo?e8qAq)ET
z2<-(#iGCRM02O+&okPd7Ycb^M#iKRA9F%6wbX1aZ&KyRadp7wX$B=I{f|AoGEc;vD
zIPu1XgBJAW#&end$K~ZH`@*?*4vlEj)5-_D^1;QO2wkOc=YBQ1{=IWQ{>N%BWM1jL
z9<)C3&9_g#<aw@o;_20a>pcWXg=h5AbtwBDEwf>F;`}A8y?r~%?z5m@{+r^sWAa%`
z_G`i6Kb=76wSjN^@}O-wLXV-z>MW%vRJy@Hi+imO{=fd$f@%{kqP)j70VBf6%>dvh
z>mPMe^&KSCQ+5zA1?468<~y=Rq%J$?l^*KdqC?&51cOJKoD<U{HK6vLUz`HP?+?(F
zdT@3`)3g&>*EL&SzJF2T^iP(oY~~7$o1=hHUwL`q6gtmS4Rz@6Xe?<3HIEIv{I!9{
z(47ab%*$Z;-tculI|_W}J;c1&4zzzl`+<GmZRiiLKYIhR=G~&Fg53dFD;%$J>TWwJ
zT)lnl;-%xidqXSqx;Wouu1TiBhyVC98)`K>i0QV`%JMVL6J_2u`gB1GBTlajNJzWC
znZcCQ9ARux?KUXTVV`YGyWl}*U(vMgEWA|2JUY->-8IpS`DhZn$hN%loo>@L0cD?0
z^rvR7^R;B@g~MhA#y;!m0|IHUw+f@vK?_>@`uf?VXW!Sf_o3lWKdb%A#p@uRxJG>?
zMbQs^5HYW)G<+}-(>MV=u^0vr&c+j?{Gi=JD%kSpGpd8LI-Q5%^)%@8C41x8F`R1{
zzXV#xw63Z4oYpwKId`Gck(1Hr2s{W~h>xFt?%2qM2Zx?}_l2?7pFel(`}aQ|JonDK
z$A_MK^<v|>3t%mG2!^;0YF76lcZm$B73axKG<xp^|Alin_s-_rI}gCc*mE27&_ehB
z`+kq`Mf_*{NBjqT9>0cvivJV;J3ft{#n0fU@Jakl{1|>1AH@&gd$5cLaTtFYcjFuJ
zHhc}<fY;(S+=T0J74~AAPxDRsa=y1fB|8`aMt~7u1Q-EEfDvE>7y(9r5nu!uflrfw
z4OZP;TN@RvtyBbqRIFG*#q#A;w6svs+)PDN6BUh(REQ!K4GmN*TSmpwrBu|{Q&Cq(
zMQtq=H8oT$Swcl1K*i$4R8&_}QB_5S-%rJ&MO0K)Qc+Ps1;$kPd{lV7RCqj8xZPB^
zTvRxnR5%<|*zHu<Y*bjSR0skUJWqwiA_z_={r&&U5AQ$CT4A;r0Y-okU<4QeMt~7u
z1Q-EEfDvE>7y(A$QzJlr|IgO{pBipvh!J1}7y(9r5nu!u0Y-okU<4QeMt~9cObF2Z
z|9Cz5`~UZF4!@1xz~lHgum|Ah_+|VO9>Ztw4`Cm`_wdvBpYW6TTljI<3-Bl&#YgZV
zJdD$@AK)%5;r)0Yz7_9<JptSC7Q7i>1uEIW2rvSS03*N%FanGKBftnS0*nA7zzBSv
z3Gi+3-h3Y3s^H~=Bv?U$<s@h!K{E-ONYF?EkpvASSVn@SB&a7r9SLelP(y+xBnXgT
zF$t<kP(=bi2^Nu{k_7PX8844X;3I*T1RfH&N#G)ZlLYXl8ZU1rfsF)K5(p&VNnjxX
IdH?_a0U))~ZvX%Q
literal 0
HcmV?d00001
diff --git a/Outputs_SeqPosteriorComparison/posterior/Z.npy b/Outputs_SeqPosteriorComparison/posterior/Z.npy
new file mode 100644
index 0000000000000000000000000000000000000000..8d89efa6714257ec2d867aa5eb95b7f23b915010
GIT binary patch
literal 176
zcmbR27wQ`j$;eQ~P_3SlTAW;@Zl$1ZlV+i=qoAIaUsO_*m=~X4l#&V(cT3DEP6dh=
qXCxM+0{I%oItoUbItsN4WCJb+G%#x>2<)_n2!F7L@J~Yd59|R^<R<I@
literal 0
HcmV?d00001
diff --git a/src/bayesvalidrox/.coverage b/src/bayesvalidrox/.coverage
new file mode 100644
index 0000000000000000000000000000000000000000..a9949086ff22941448ad6b80430111e3dd99e24f
GIT binary patch
literal 53248
zcmeI)O>f&a7zc1EZmrl!;0{3`1U`h9%`s}bFmf2S9ky)Cp;&>gTd>Q3xwK3t>MY68
zOI{Ac;C4WO0s98SzSF+OE<NqE%brJ+EIWyd-U_IHA$IgaQsj>hNxN$M{j(=-ti(77
zJsFDw<AGtC#uq{uhOthsEqdjfMLYNM1Nv<)?62BgH+tiLwi-Veo7F#z#!p*M8{N%6
zw?5kVd(*4^zVVkOYYtt&1_1~_00Qr~!0g$kW$*5qFMo;UaH8T+Ix5ufTVH*5ba;Fu
zjt{?lawPOQv0oRob-UtFgu#WFs!)vGi4v|qavd4F{)vcBRh~MD)QA^4I!8+#C+wy9
zd2is3C|9gbsEBFkdNRBcKdY-wnjpJdtS)1Hf(lXYiO(s-ewMv0LN!*Q@*NfFTI{={
zcD?=T*SckIZ=2UDZIcWg`q@}&4V@^O@_0-IIYBt$+7CnNJEtn@2sv+|6Hq-bW8K;?
z@E*m1n7Dr0Oh0mCH}HkJR8A7B(VY$KFj~G5-ePO0tmoXI=Q?M0uLYeV^Bp;}<XPr1
zRn8A`)*>U{nlh(#_7iVN<S!C$%AG?w^o`?SRnAp;;~;D7TldW!eOa`3sXKFW?opj9
z@!hjT%{#v%ikzLa8Gg_a+z5(*4>zjzC)=h$QK>GYvk86P8ptFL(&GVj{-959o4<Xq
zZrOW#=Ji<`D9ke$x^Xn$sVxP~{Pc>@;p?m8hDXVAknkv)OhhnFhYaTyx^pO+7st$_
z97BdY&_s|Wq?*3dXyeLK9mH@F43}dqjhIN$0HL`cPc}|`hlfhUZE%8f70MIU%dYBa
z;avw3g-*X#wU72zh0Z(#2L1VNeYRV*>}Jz^Sxdtx>#gjUwG={mPu?+&=9fGrqa@5H
zk~d9~d5&V1%*RM}g__f~s`g=XRdcd*gMPNNKC|vw_Rfwu(<6fCC4SvsY6PDthQN)5
z;O(nFmGP;#U3WfB6>r_I+Mn&LsyJ6+(C7E+s~^|WDqruW`W{XC{HiQfna>ne{zhGO
z&*g+>lc{t=8nQG8>lHw7!jYuJau_6W8ozv)h9|E*MSXPv#c^&=soH6B8s6UqQRLFH
z%F9;tX?|9;PG!WiP}CB`AeboW>$QP$JdWs|ZY0t=++LMFFUp9nxQvtYR$8Zgap7tH
zm<C-wzG%nuq(92mD_(|msWkWV;N<)0NrW<r77faXhUkql^Fp($JY(hM_oigRrFwFt
zhdX!5LWvir>Q-@jHTqh{Dt0|3I7wQWG_WpYB%DyuMOsVSt;(CifO~7u=Xd^=%-^1k
z*`8_H4<DMdBpZx+o*6{6sCzQ6c6PY76eKq;7V+|wCS3Zovlu6Ax4^-N`VKs~=^h+v
zg*;XES*PCSFZgETKZAa-K>z{}fB*y_009U<00Izz00baVK7oq4W?K6D|7)Z1x_k{p
ziy#022tWV=5P$##AOHafKmY>osX)W5H0|s!0esv$=pEDv!fgC&&<{2UKmY;|fB*y_
z009U<00Izz00hb;P_Hy=*&hO=|G@u2nHq<-KmY;|fB*y_009U<00Izz00ba#Q-GiU
z(|`Jh4FV8=00bZa0SG_<0uX=z1Rwx`5)0t@e~CL6O@jagAOHafKmY;|fB*y_009VO
z0(kz9P=EjgAOHafKmY;|fB*y_009V;UjWbl%iqUnAp{@*0SG_<0uX=z1Rwwb2tWYO
z{}BTafB*y_009U<00Izz00bZaf$|IB`G5KQ7%hYV1Rwwb2tWV=5P$##AOHaf;Q2pd
z00Izz00bZa0SG_<0uX=z1Rzj;0X+XNe;=cT5P$##AOHafKmY;|fB*y_0D=DjZpReE
literal 0
HcmV?d00001
diff --git a/src/bayesvalidrox/surrogate_models/input_space.py b/src/bayesvalidrox/surrogate_models/input_space.py
index 99b0d5d39..7eedbbe0d 100644
--- a/src/bayesvalidrox/surrogate_models/input_space.py
+++ b/src/bayesvalidrox/surrogate_models/input_space.py
@@ -30,7 +30,7 @@ class InputSpace:
self.meta_model_type = meta_model_type
# Other
- self.apce = None
+ self.pce = None
self.bound_tuples = None
self.input_data_given = None
self.JDist = None
@@ -61,12 +61,11 @@ 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':
- self.apce = True
+ # Check if PCE metamodel is selected.
+ if self.meta_model_type.lower() == 'pce':
+ self.pce = True
else:
- self.apce = False
+ self.pce = False
# check if marginals given
if not self.ndim >= 1:
@@ -293,7 +292,7 @@ class InputSpace:
f"{parIdx + 1} is not available.")
raise ValueError(message)
- if self.input_data_given or self.apce:
+ if self.input_data_given or not self.pce:
polytype = 'arbitrary'
# Store dists and poly_types
@@ -376,7 +375,7 @@ class InputSpace:
disttypes.append(inputs.Marginals[par_i].dist_type)
# Pass non-transformed X, if arbitrary PCE is selected.
- if None in disttypes or self.input_data_given or self.apce:
+ if None in disttypes or self.input_data_given or not self.pce:
return X
cdfx = np.zeros(X.shape)
diff --git a/tests/test_InputSpace.py b/tests/test_InputSpace.py
index 40f269563..60893db8f 100644
--- a/tests/test_InputSpace.py
+++ b/tests/test_InputSpace.py
@@ -125,7 +125,7 @@ def test_check_valid_input_noapc() -> None:
inp.add_marginals()
inp.Marginals[1].dist_type = 'normal'
inp.Marginals[1].parameters = [0, 1]
- InputSpace(inp, meta_Model_type='gpe')
+ InputSpace(inp, meta_model_type='gpe')
#%% Test ExpDesign.build_polytypes
diff --git a/tests/test_MetaModel.py b/tests/test_MetaModel.py
index 5cea0dcae..dd1eb29f2 100644
--- a/tests/test_MetaModel.py
+++ b/tests/test_MetaModel.py
@@ -33,8 +33,8 @@ sys.path.append("src/")
from bayesvalidrox.surrogate_models.inputs import Input
from bayesvalidrox.surrogate_models.input_space import InputSpace
-from bayesvalidrox.surrogate_models.surrogate_models import MetaModel, corr_loocv_error, create_psi
-from bayesvalidrox.surrogate_models.surrogate_models import gaussian_process_emulator
+from bayesvalidrox.surrogate_models.surrogate_models import MetaModel
+from bayesvalidrox.surrogate_models.supplementary import create_psi
@@ -51,137 +51,31 @@ def test_metamod() -> None:
MetaModel(inp)
-#%% Test MetaModel.build_metamodel
+#%% Test MetaModel.check_is_gaussian
-def test_build_metamodel_nosamples() -> None:
- """
- Build MetaModel without collocation samples
- """
+def test_check_is_gaussian() -> None:
inp = Input()
inp.add_marginals()
inp.Marginals[0].dist_type = 'normal'
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:
- """
- Build MetaModel
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- mm.CollocationPoints = np.array([[0.2], [0.8]])
- mm.build_metamodel()
-
+ mm.check_is_gaussian()
-def test_build_metamodel_ninitsamples() -> None:
- """
- Build MetaModel with n_init_samples
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- mm.CollocationPoints = np.array([[0.2], [0.8]])
- mm.build_metamodel(n_init_samples=2)
+#%% Test MetaModel.build_metamodel
-def test_build_metamodel_gpe() -> None:
+def test_build_metamodel() -> None:
"""
- Build MetaModel gpe
+ Build 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.CollocationPoints = np.array([[0.2], [0.8]])
mm.build_metamodel()
-
-
-def test_build_metamodel_coldimerr() -> None:
- """
- Build MetaModel with wrong shape collocation samples
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- mm.CollocationPoints = [[0.2, 0.8]]
- with pytest.raises(AttributeError) as excinfo:
- mm.build_metamodel()
- 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:
- """
- Generate polynomials without ExpDeg
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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:
- """
- Generate polynomials with max_deg
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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
@@ -261,656 +155,6 @@ def test_fit_gpe() -> None:
mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
-#%% Test MetaModel.create_psi
-
-def test_create_psi() -> None:
- """
- Create psi-matrix
- """
- 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], [0.8]])
- 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)
- create_psi(BasisIndices, univ_bas)
-
-
-#%% Test MetaModel.regression
-
-def test_regression() -> None:
- """
- Regression without a method
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi)
-
-
-def test_regression_ols() -> None:
- """
- Regression: ols
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='ols')
-
-
-def test_regression_olssparse() -> None:
- """
- Regression: ols and sparse
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='ols', sparsity=True)
-
-
-def test_regression_ard() -> None:
- """
- Regression: ard
- """
- 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], [0.8]])
- outputs = np.array([0.4, 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='ard')
-
-
-def test_regression_ardssparse() -> None:
- """
- Regression: ard and sparse
- """
- 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], [0.8]])
- outputs = np.array([0.4, 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='ard', sparsity=True)
-
-
-def test_regression_fastard() -> None:
- """
- Regression: fastard
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='fastard')
-
-
-def test_regression_fastardssparse() -> None:
- """
- Regression: fastard and sparse
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='fastard', sparsity=True)
-
-
-def test_regression_brr() -> None:
- """
- Regression: brr
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='brr')
-
-
-def test_regression_brrssparse() -> None:
- """
- Regression: brr and sparse
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='brr', sparsity=True)
-
-
-if 0: # Could not figure out these errors, issue most likely in chosen samples/outputs
- def test_regression_bcs() -> None:
- """
- Regression: bcs
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
- mm.pce_deg = 3
- mm.CollocationPoints = samples
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='bcs')
-
-
- def test_regression_bcsssparse() -> None:
- """
- Regression: bcs and sparse
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
-
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='bcs', sparsity=True)
-
-
-def test_regression_lars() -> None:
- """
- Regression: lars
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
-
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='lars')
-
-
-def test_regression_larsssparse() -> None:
- """
- Regression: lars and sparse
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
-
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='lars', sparsity=True)
-
-
-def test_regression_sgdr() -> None:
- """
- Regression: sgdr
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='sgdr')
-
-
-def test_regression_sgdrssparse() -> None:
- """
- Regression: sgdr and sparse
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='sgdr', sparsity=True)
-
-
-if 0: # Could not figure out these errors, issue most likely in chosen samples/outputs
- def test_regression_omp() -> None:
- """
- Regression: omp
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1])
-
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='omp')
-
-
- def test_regression_ompssparse() -> None:
- """
- Regression: omp and sparse
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='omp', sparsity=True)
-
-
-def test_regression_vbl() -> None:
- """
- Regression: vbl
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='vbl')
-
-
-def test_regression_vblssparse() -> None:
- """
- Regression: vbl and sparse
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='vbl', sparsity=True)
-
-
-def test_regression_ebl() -> None:
- """
- Regression: ebl
- """
- 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 = create_psi(BasisIndices, univ_bas)
-
- mm.regression(samples, outputs, psi, reg_method='ebl')
-
-
-def test_regression_eblssparse() -> None:
- """
- Regression: ebl and sparse
- """
- 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 = 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...
-
-#%% Test MetaModel.univ_basis_vals
-
-def test_univ_basis_vals() -> None:
- """
- Creates univariate polynomials
- """
- 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], [0.8]])
- mm.CollocationPoints = samples
- 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
- """
- 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([0.8])
-
- mm.CollocationPoints = samples
- 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)
-
- with pytest.raises(AttributeError) as excinfo:
- 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
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- samples = np.array([[0.0], [0.1]])
- outputs = np.array([0.0, 0.1])
-
- mm.CollocationPoints = samples
- 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(outputs, 0)
-
-
-def test_adaptive_regression_verbose() -> None:
- """
- Adaptive regression, no specific method, verbose output
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- samples = np.array([[0.0], [0.1]])
- outputs = np.array([0.0, 0.1])
-
- mm.CollocationPoints = samples
- 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(outputs, 0, True)
-
-
-def test_adaptive_regression_ols() -> None:
- """
- Adaptive regression, ols
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
-
- mm.CollocationPoints = samples
- 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.pce_reg_method = 'ols'
- mm.adaptive_regression(outputs, 0)
-
-
-#%% Test MetaModel.corr_loocv_error
-
-def test_corr_loocv_error_nosparse() -> None:
- """
- Corrected loocv error
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
-
- 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 = 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
- """
- 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([0.1])
-
- 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 = 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
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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])
-
- 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 = 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
@@ -924,7 +168,7 @@ def test_pca_transformation() -> None:
inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
- mm.pca_transformation(outputs)
+ mm.pca_transformation(outputs, 1)
def test_pca_transformation_varcomp() -> None:
@@ -938,7 +182,7 @@ def test_pca_transformation_varcomp() -> None:
mm = MetaModel(inp)
outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
mm.var_pca_threshold = 1
- mm.pca_transformation(outputs)
+ mm.pca_transformation(outputs, 1)
def test_pca_transformation_ncomp() -> None:
@@ -950,55 +194,8 @@ def test_pca_transformation_ncomp() -> None:
inp.Marginals[0].dist_type = 'normal'
inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
- mm.n_pca_components = 1
- mm.pca_transformation(outputs)
-
-
-#%% Test MetaModel.gaussian_process_emulator
-
-def test_gaussian_process_emulator() -> None:
- """
- Create GPE
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5])
-
-
-def test_gaussian_process_emulator_nug() -> None:
- """
- Create GPEwith nugget
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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:
- """
- Fit MetaModel with autoselect
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- 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:
- """
- Create GPE with var_idx
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5], varIdx=1)
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
+ mm.pca_transformation(outputs, 1)
#%% Test MetaModel.eval_metamodel
@@ -1012,70 +209,10 @@ def test_eval_metamodel() -> None:
inp.Marginals[0].dist_type = 'normal'
inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
- mm.out_names = ['Z']
- mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
- mm.eval_metamodel([[0.4]])
-
-
-def test_eval_metamodel_normalboots() -> None:
- """
- Eval trained MetaModel with normal bootstrap
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- mm.bootstrap_method = 'normal'
- mm.out_names = ['Z']
- mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
- mm.eval_metamodel([[0.4]])
-
-
-def test_eval_metamodel_highnormalboots() -> None:
- """
- Eval trained MetaModel with higher bootstrap-itrs
- """
- inp = Input()
- inp.add_marginals()
- inp.Marginals[0].dist_type = 'normal'
- inp.Marginals[0].parameters = [0, 1]
- mm = MetaModel(inp)
- mm.n_bootstrap_itrs = 2
- mm.out_names = ['Z']
- mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
- mm.eval_metamodel([[0.4]])
-
-
-def test_eval_metamodel_gpe() -> None:
- """
- Eval trained MetaModel - gpe
- """
- 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.out_names = ['Z']
- mm.fit([[0.2], [0.8]], {'Z': np.array([[0.4], [0.5]])})
- mm.eval_metamodel([[0.4]])
+ out = mm.eval_metamodel([[0.4]])
+ assert len(out) == 2
-def test_eval_metamodel_pca() -> None:
- """
- Eval trained MetaModel 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.out_names = ['Z']
- 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
@@ -1115,7 +252,7 @@ def test_copy_meta_model_opts() -> None:
#%% Test Engine._compute_pce_moments
-def test__compute_pce_moments() -> None:
+def test__compute_moments() -> None:
"""
Compute moments of a pce-surrogate
"""
@@ -1125,37 +262,9 @@ def test__compute_pce_moments() -> None:
inp.Marginals[0].parameters = [0, 1]
mm = MetaModel(inp)
mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
- mm._compute_pce_moments()
+ mm._compute_moments()
-def test__compute_pce_moments_pca() -> None:
- """
- Compute moments of a pce-surrogate 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]]})
- mm._compute_pce_moments()
-
-
-def test__compute_pce_moments_gpe() -> None:
- """
- Compute moments of a gpe-surrogate
- """
- 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'
- 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
diff --git a/tests/test_PolynomialChaosEmulator.py b/tests/test_PolynomialChaosEmulator.py
new file mode 100644
index 000000000..c15f9e139
--- /dev/null
+++ b/tests/test_PolynomialChaosEmulator.py
@@ -0,0 +1,952 @@
+# -*- coding: utf-8 -*-
+"""
+Test the MetaModel class in bayesvalidrox.
+Tests are available for the following functions
+Class MetaModel:
+ build_metamodel - x
+ update_metamodel
+ update_pce_coeffs
+ create_basis_indices --removed, just redirects
+ add_InputSpace -x
+ univ_basis_vals
+ fit
+ adaptive_regression
+ pca_transformation
+ eval_metamodel
+ create_model_error
+ eval_model_error
+ auto_vivification
+ copy_meta_model_opts
+ __select_degree
+ generate_polynomials
+ _compute_pce_moments
+
+"""
+import numpy as np
+import pytest
+import sys
+
+sys.path.append("src/")
+
+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.polynomial_chaos import PCE
+from bayesvalidrox.surrogate_models.supplementary import gaussian_process_emulator, corr_loocv_error, create_psi
+
+
+
+#%% Test MetaMod constructor on its own
+
+def test_metamod() -> None:
+ """
+ Construct PCE without inputs
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ PCE(inp)
+
+
+#%% Test PCE.build_metamodel
+
+def test_build_metamodel_nosamples() -> None:
+ """
+ Build PCE without collocation samples
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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:
+ """
+ Build PCE
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.CollocationPoints = np.array([[0.2], [0.8]])
+ mm.build_metamodel()
+
+
+def test_build_metamodel_ninitsamples() -> None:
+ """
+ Build PCE with n_init_samples
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.CollocationPoints = np.array([[0.2], [0.8]])
+ mm.build_metamodel(n_init_samples=2)
+
+
+
+def test_build_metamodel_coldimerr() -> None:
+ """
+ Build PCE with wrong shape collocation samples
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.CollocationPoints = [[0.2, 0.8]]
+ with pytest.raises(AttributeError) as excinfo:
+ mm.build_metamodel()
+ 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:
+ """
+ Generate polynomials without ExpDeg
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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) == 'PCE cannot generate polynomials in the given scenario!'
+
+
+def test_generate_polynomials_deg() -> None:
+ """
+ Generate polynomials with max_deg
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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:
+ """
+ Add InputSpace in PCE
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.add_InputSpace()
+
+
+#%% Test PCE.fit
+# Faster without these
+def test_fit() -> None:
+ """
+ Fit PCE
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
+
+
+def test_fit_parallel() -> None:
+ """
+ Fit PCE in parallel
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]}, parallel=True)
+
+
+def test_fit_verbose() -> None:
+ """
+ Fit PCE verbose
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]}, verbose=True)
+
+
+def test_fit_pca() -> None:
+ """
+ Fit PCE verbose and with pca
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.dim_red_method = 'pca'
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]}, verbose=True)
+
+
+
+#%% Test PCE.regression
+
+def test_regression() -> None:
+ """
+ Regression without a method
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi)
+
+
+def test_regression_ols() -> None:
+ """
+ Regression: ols
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ols')
+
+
+def test_regression_olssparse() -> None:
+ """
+ Regression: ols and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ols', sparsity=True)
+
+
+def test_regression_ard() -> None:
+ """
+ Regression: ard
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ samples = np.array([[0.2], [0.8]])
+ outputs = np.array([0.4, 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ard')
+
+
+def test_regression_ardssparse() -> None:
+ """
+ Regression: ard and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ samples = np.array([[0.2], [0.8]])
+ outputs = np.array([0.4, 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ard', sparsity=True)
+
+
+def test_regression_fastard() -> None:
+ """
+ Regression: fastard
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='fastard')
+
+
+def test_regression_fastardssparse() -> None:
+ """
+ Regression: fastard and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='fastard', sparsity=True)
+
+
+def test_regression_brr() -> None:
+ """
+ Regression: brr
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='brr')
+
+
+def test_regression_brrssparse() -> None:
+ """
+ Regression: brr and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='brr', sparsity=True)
+
+
+if 0: # Could not figure out these errors, issue most likely in chosen samples/outputs
+ def test_regression_bcs() -> None:
+ """
+ Regression: bcs
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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])
+ mm.pce_deg = 3
+ mm.CollocationPoints = samples
+ 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='bcs')
+
+
+ def test_regression_bcsssparse() -> None:
+ """
+ Regression: bcs and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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])
+
+ 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='bcs', sparsity=True)
+
+
+def test_regression_lars() -> None:
+ """
+ Regression: lars
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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])
+
+ 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='lars')
+
+
+def test_regression_larsssparse() -> None:
+ """
+ Regression: lars and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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])
+
+ 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='lars', sparsity=True)
+
+
+def test_regression_sgdr() -> None:
+ """
+ Regression: sgdr
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='sgdr')
+
+
+def test_regression_sgdrssparse() -> None:
+ """
+ Regression: sgdr and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='sgdr', sparsity=True)
+
+
+if 0: # Could not figure out these errors, issue most likely in chosen samples/outputs
+ def test_regression_omp() -> None:
+ """
+ Regression: omp
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.1])
+
+ 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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='omp')
+
+
+ def test_regression_ompssparse() -> None:
+ """
+ Regression: omp and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='omp', sparsity=True)
+
+
+def test_regression_vbl() -> None:
+ """
+ Regression: vbl
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='vbl')
+
+
+def test_regression_vblssparse() -> None:
+ """
+ Regression: vbl and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='vbl', sparsity=True)
+
+
+def test_regression_ebl() -> None:
+ """
+ Regression: ebl
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = create_psi(BasisIndices, univ_bas)
+
+ mm.regression(samples, outputs, psi, reg_method='ebl')
+
+
+def test_regression_eblssparse() -> None:
+ """
+ Regression: ebl and sparse
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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 = 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...
+
+#%% Test PCE.univ_basis_vals
+
+def test_univ_basis_vals() -> None:
+ """
+ Creates univariate polynomials
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ samples = np.array([[0.2], [0.8]])
+ mm.CollocationPoints = samples
+ mm.build_metamodel(n_init_samples=2)
+ mm.univ_basis_vals(samples)
+
+
+#%% Test PCE.adaptive_regression
+
+def test_adaptive_regression_fewsamples() -> None:
+ """
+ Adaptive regression, no specific method, too few samples given
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ samples = np.array([[0.2]])
+ outputs = np.array([0.8])
+
+ mm.CollocationPoints = samples
+ 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)
+
+ with pytest.raises(AttributeError) as excinfo:
+ 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
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ samples = np.array([[0.0], [0.1]])
+ outputs = np.array([0.0, 0.1])
+
+ mm.CollocationPoints = samples
+ 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(outputs, 0)
+
+
+def test_adaptive_regression_verbose() -> None:
+ """
+ Adaptive regression, no specific method, verbose output
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ samples = np.array([[0.0], [0.1]])
+ outputs = np.array([0.0, 0.1])
+
+ mm.CollocationPoints = samples
+ 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(outputs, 0, True)
+
+
+def test_adaptive_regression_ols() -> None:
+ """
+ Adaptive regression, ols
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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])
+
+ mm.CollocationPoints = samples
+ 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.pce_reg_method = 'ols'
+ mm.adaptive_regression(outputs, 0)
+
+
+#%% Test PCE.pca_transformation
+
+def test_pca_transformation() -> None:
+ """
+ Apply PCA
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
+ mm.pca_transformation(outputs, 1)
+
+
+def test_pca_transformation_varcomp() -> None:
+ """
+ Apply PCA with set var_pca_threshold
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ outputs = np.array([[0.4, 0.4], [0.5, 0.6]])
+ mm.var_pca_threshold = 1
+ mm.pca_transformation(outputs, 1)
+
+
+
+#%% Test PCE.eval_metamodel
+
+def test_eval_metamodel() -> None:
+ """
+ Eval trained PCE
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.out_names = ['Z']
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
+ mm.eval_metamodel([[0.4]])
+
+
+def test_eval_metamodel_normalboots() -> None:
+ """
+ Eval trained PCE with normal bootstrap
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.bootstrap_method = 'normal'
+ mm.out_names = ['Z']
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
+ mm.eval_metamodel([[0.4]])
+
+
+def test_eval_metamodel_highnormalboots() -> None:
+ """
+ Eval trained PCE with higher bootstrap-itrs
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.n_bootstrap_itrs = 2
+ mm.out_names = ['Z']
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
+ mm.eval_metamodel([[0.4]])
+
+
+def test_eval_metamodel_pca() -> None:
+ """
+ Eval trained PCE with pca
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(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.eval_metamodel([[0.4]])
+
+
+#%% Test PCE.create_model_error
+# TODO: move model out of this function
+
+#%% Test PCE.eval_model_error
+# TODO: test create_model_error first
+
+#%% Test PCE.auto_vivification
+def test_auto_vivification() -> None:
+ """
+ Creation of auto-vivification objects
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.auto_vivification()
+
+
+#%% Test PCE.copy_meta_model_opts
+
+def test_copy_meta_model_opts() -> None:
+ """
+ Copy the PCE with just some stats
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.add_InputSpace()
+ mm.copy_meta_model_opts()
+
+
+#%% Test PCE.__select_degree
+
+#%% Test Engine._compute_moments
+
+def test__compute_moments() -> None:
+ """
+ Compute moments of a pce-surrogate
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.fit([[0.2], [0.4], [0.8]], {'Z': [[0.4], [0.2], [0.5]]})
+ mm._compute_moments()
+
+
+def test__compute_moments_pca() -> None:
+ """
+ Compute moments of a pce-surrogate with pca
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ mm = PCE(inp)
+ mm.dim_red_method = 'pca'
+ mm.fit([[0.2], [0.8]], {'Z': [[0.4, 0.4], [0.5, 0.6]]})
+ mm._compute_moments()
+
+
+#%% Test PCE.update_metamodel
+# TODO: taken from engine
diff --git a/tests/test_supplementary.py b/tests/test_supplementary.py
new file mode 100644
index 000000000..7537b0ca3
--- /dev/null
+++ b/tests/test_supplementary.py
@@ -0,0 +1,162 @@
+# -*- coding: utf-8 -*-
+"""
+Test the supplementary functions in bayesvalidrox.
+Tests are available for the following functions
+ hellinger_distance
+ logpdf
+ subdomain
+ create_psi x
+ corr_loocv_error x
+ gaussian_process_emulator x
+
+"""
+import numpy as np
+import pytest
+import sys
+
+sys.path.append("src/")
+
+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.polynomial_chaos import PCE
+from bayesvalidrox.surrogate_models.supplementary import gaussian_process_emulator, corr_loocv_error, create_psi
+
+
+#%% Test create_psi
+
+def test_create_psi() -> None:
+ """
+ Create psi-matrix
+ """
+ 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], [0.8]])
+ 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)
+ create_psi(BasisIndices, univ_bas)
+
+#%% Test corr_loocv_error
+
+def test_corr_loocv_error_nosparse() -> None:
+ """
+ Corrected loocv error
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ 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])
+
+ 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 = 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
+ """
+ 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([0.1])
+
+ 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 = 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
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ 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])
+
+ 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 = 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 gaussian_process_emulator
+
+def test_gaussian_process_emulator() -> None:
+ """
+ Create GPE
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5])
+
+
+def test_gaussian_process_emulator_nug() -> None:
+ """
+ Create GPEwith nugget
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ 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:
+ """
+ Fit MetaModel with autoselect
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ 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:
+ """
+ Create GPE with var_idx
+ """
+ inp = Input()
+ inp.add_marginals()
+ inp.Marginals[0].dist_type = 'normal'
+ inp.Marginals[0].parameters = [0, 1]
+ gaussian_process_emulator([[0.2], [0.8]], [0.4, 0.5], varIdx=1)
+
--
GitLab