>h\ SSKr"SS5rg)Nc\rSrSrSrSrSrSSjrSrSr S r S r S r SS jr SS jrSrSrSSjrSSjrSSjrSrg)Pcaz| A basic class for Principal Component Analysis (PCA). p is the number of dimensions, while N is the number of data points )rgbcymkcURnU[R"USS9- nU[R"USS9- nX lX0lSUlg)Nraxis)AnpmeanstdMN_eig)selfrrrs jC:\Users\julio\OneDrive\Documentos\Trabajo\Ideas Frescas\venv\Lib\site-packages\statsmodels/sandbox/pca.py__calc Pca.__calc sF FF BGGA1   BFF1!   Nc L[R"U5RnURupEXEsUlUlXT:aSSKJn U"S[5 X0l UR5Ul UR5 [R"UR[US- [!UR5- 5S-5SUUlUb[!U5U:wa [#S5eUcSUlg[%UVs/sHn['U5PM sn5Ulgs snf)z p X N matrix input r)warnzp > n - intentional?Nznames must match data dimension)rarrayTshapenpwarningsrRuntimeWarningrcopy_origA _Pca__calctile_colorsintlen ValueErrortuplestrnames)rdatar0rr"r#rxs r__init__ Pca.__init__s HHTN  gg tv 5 % ' 8FFH  ggdll3!S5F/F+G+IJ2AN  Uq>? ?"]T u6Mu!s1vu6M0N 6Ms?D!cV[R"URR5$)z/ returns the covariance matrix for the dataset )rcovrr rs rgetCovarianceMatrixPca.getCovarianceMatrix+svvdffhhrcURcf[RRUR 55n[R "US5SSS2nUSUUSSS2U44nXlUR$)zA returns a tuple of (eigenvalues,eigenvectors) for the data set. Nrr)rrlinalgeigr8argsort)rressortis rgetEigensystemPca.getEigensystem1sq 99 ))-- 8 8 :;C**SV$TrT*EQs1vag/CIyyrc(UR5S$)NrrAr7s rgetEigenvaluesPca.getEigenvalues<""$Q''rc(UR5S$)NrrDr7s rgetEigenvectorsPca.getEigenvectors?rGrcTUR5nU[R"U5- $)z- "energies" are just normalized eigenvectors )rErsum)rvs r getEnergiesPca.getEnergiesBs$    {rc ~SSKJn URSS2U4URSS2U4peU(aUR5 UR XV5 UR 5upxUR 5upUR5upX- X- p[XxRUR5H3unnnURSSUUU-UUU-SX-S- S--UUS9 M5 URbIURSURU-S-5 URSURU-S-5 gg) z Generates a 2-dimensional plot of the data set and principle components using matplotlib. ix specifies which p-dimension to put on the x-axis of the plot and iy specifies which to put on the y-axis (0-indexed) rNg?g?) head_widthfcec$z/\sigma$)matplotlib.pyplotpyplotrclfscatterrAxlimylimzipr r*arrowr0xlabelylabel)rixiyrXpltr2r valsevsxlxuylyudxdyvalvecr s rplot2d Pca.plot2dIs ( FF1R4L"!  GGI A$$&hhjhhjuru2T%% 5ICA IIa#c"g+c#b'kT257S.=PTUYZI [6 :: ! JJs4::b>)+5 6 JJs4::b>)+5 6 "rc 6SSKJs Jn U(aUR5 [R "S5nUR 5UR5-nURXfXgUXrXsSS9 URURSS2U4URSS2U4URSS2U4SS9 UR(aBURURUS-URUS-URUS-S9 gUR5 g) z Generates a 3-dimensional plot of the data set and principle components using mayavi. ix, iy, and iz specify which of the input p-dimensions to place on each of the x,y,z axes, respectively (0-indexed). rNr) scale_factorg333333?z/sigma)r^r_zlabel) enthought.mayavi.mlabmayavimlabrXrzerosrIrEquiver3dpoints3drr0axes)rr`raizrXrz3rMs rplot3d Pca.plot3dcs *)  EEG 88A;    !$"5"5"7 7 2bE!%1 = 466!B$<qt TVVAbD\s K :: FF$**R.1B9PX\XbXbceXfgoXoF p FFHrc[R"U5(a0U[R"URRS5-nU[R "URSS9-nURRSn[R "[R"UR5U:SS9nURUUlUR5 U[U5- $)z clips out all data points that are more than a certain number of standard deviations from the mean. sigs can be either a single value or a length-p sequence that specifies the number of standard deviations along each of the p dimensions. rrr) risscalaronesrr!rallabsrr(rL)rsigsr"r s rsigclip Pca.sigclipws ;;t  bggdffll1o..DBFF466q)) FFLLO FF266$&&>D(a 0vvay QxrcbURR5UlUR5 gN)r'r&rr(r7s rreset Pca.resets!!# rc[X#U4Vs/sHoUSLPM sn5nUS:Xa [S5nOnUS:a [S5eUbUR5U:nOFUb [SU5nO6Ub([R "UR55U:nO [ S5eUcURRnOT[R"USS9nURRRSURS:wa [S5e[R"UR55RU-nXR$s snf) a projects the normalized values onto the components enthresh, nPCs, and cumen determine how many PCs to use if vals is None, the normalized data vectors are the values to project. Otherwise, it should be convertable to a p x N array returns n,p(>threshold) dimension array Nrrz&cannot specify more than one thresholdzShould be unreachableF)r&zshape for vals does not match) rLslicer-energiesrcumsum RuntimeErrorrr rr!matrixrI) rrcenthreshnPCscumenenonnonesr projs rproject Pca.projects/FG/F! /FGH q=d A \EF F#MMOh.!$t$"IIdmmo.%7"#:;; <6688D88De,Dvvxx~~a DJJqM1 !@AAyy--/02247wyy-Hs E c[R"U5nURup4URRSnXE:a [ S5e[R R [R"UR55R5n[R"X545nXSS2SU24'XgR-nU(a*[R"UR5R$[R"URSS9n [R"URSS9n [R"UR5U -U -R$)z8 input is an n X q array, where q <= p output is p X n rzq > pNrr)r atleast_2dr!rr-r<invrrIr rvrrrr) rrnormedr"qr#evinvzsrmnssdss r deproject Pca.deprojects -- gg FFLLO 5W% %iimmBIId&:&:&<=??@ XXqe_1RaR4TTz 88DFF#%% %A&Ctvv1%CHHTVV$S(,// /rcUc URnOAURnURSURRS:wa [S5eUR 5n[ R "U5nUSS2U4USS2U4'URUS5nURU- nUR[ R"URSS9-nU[ R"URSS9-$)z pc can be a scalar or any sequence of pc indecies if vals is None, the source data is self.A, else whatever is in vals (which must be p x m) Nrz1vals do not have the correct number of componentsFrr) rr r!r-rr zeros_likerrrr)rpcrcpcszpcsupcrBs r subtractPCPca.subtractPCs <66D66Dzz!}tvv||A. !TUU LLN ]]3 qt9QrT NN4 & FF3J CCtvv1% %Q'''r) rrrr*rr'r"r0r#r)rrT)rrT)NNNN)T)__name__ __module__ __qualname____firstlineno____doc__r*r(r3r8rArErIrNrmr|rrrrr__static_attributes__rrrrsV *GO* ((74($ !F04(rr)numpyrrrrrrs[([(r