M/Ph|(ddlZGddZdS)NcveZdZdZdZdZddZdZdZdZ d Z d Z ddZ ddZ dZdZddZddZddZdS)Pcaz A basic class for Principal Component Analysis (PCA). p is the number of dimensions, while N is the number of data points )rgbcymkc|j}|tj|dz }|tj|dz }||_||_d|_dS)Nraxis)AnpmeanstdMN_eig)selfrrrs W/var/www/html/test/jupyter/venv/lib/python3.11/site-packages/statsmodels/sandbox/pca.py__calcz Pca.__calc sT F BGA1     BF1!     Nc 6tj|j}|j\}}||c|_|_||krddlm}|dt||_ | |_ | tj |jt|dz t!|jz dzd||_|"t!||krt#d|dnt%d|D|_dS)z$ p X N matrix input r)warnzp > n - intentional?Nznames must match data dimensionc,g|]}t|S)str).0xs r z Pca.__init__..(s6M6M6M!s1vv6M6M6Mr)rarrayTshapenpwarningsrRuntimeWarningrcopy_origA _Pca__calctile_colorsintlen ValueErrortuplenames)rdatar3rr&r'rs r__init__z Pca.__init__s  HTNN g!! tv q55 % % % % % % D' 8 8 8FFHH  gdl3!S5F5F/F+G+G+IJJ2A2N  Uq>?? ?"]TT6M6Mu6M6M6M0N0N rc>tj|jjS)z? returns the covariance matrix for the dataset )rcovrr$rs rgetCovarianceMatrixzPca.getCovarianceMatrix+svdfhrc|j{tj|}tj|dddd}|d||ddd|ff}||_|jS)zQ returns a tuple of (eigenvalues,eigenvectors) for the data set. Nrr)rrlinalgeigr9argsort)rressortis rgetEigensystemzPca.getEigensystem1sz 9 )-- 8 8 : :;;C*SV$$TTrT*EQs1vaaag/CDIyrc6|dS)NrrAr8s rgetEigenvalueszPca.getEigenvalues<""$$Q''rc6|dS)NrrCr8s rgetEigenvectorszPca.getEigenvectors?rErcX|}|tj|z S)z= "energies" are just normalized eigenvectors )rDrsum)rvs r getEnergieszPca.getEnergiesBs'     {rrrTc ddlm}|jdd|f|jdd|f}}|r|||||\}}|\} } |\} } | | z | | z }} t||j |j D]@\}}}| dd|||z|||zd| |zdz dzz||A|j N| d|j |zdz|d|j |zdzdSdS) 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.arrowr3xlabelylabel)rixiyrTpltr!r valsevsxlxuylyudxdyvalvecrs rplot2dz Pca.plot2dIsu (''''' F111R4L"!   GGIII Aa$$&&Shhjj2hhjj2"ur"u2T#% 55 \ \ICA IIa#c"g+c#b'kT2b57S.=PTUYZI [ [ [ [ : ! JJs4:b>)+5 6 6 6 JJs4:b>)+5 6 6 6 6 6 " !rc dddlmcm}|r|t jd}||z}||||||||||d| |j dd|f|j dd|f|j dd|fd|j rD| |j |dz|j |dz|j |dzdS| dS) z Generates a 3-dimensional plot of the data set and principle components using mayavi. ix, iy, and iz specify which of the input p-dimensions to place on each of the x,y,z axes, respectively (0-indexed). rN) scale_factorg333333?z/sigma)rZr[zlabel) enthought.mayavi.mlabmayavimlabrTrzerosrGrDquiver3dpoints3drr3axes)rr\r]izrTrz3rJs rplot3dz Pca.plot3dcs2 *))))))))   EEGGG 8A;;    ! !$"5"5"7"7 7 2bAbE!B%"1 === 46!!!B$<qqqt TVAAAbD\s KKK :  FF$*R.1B9PX\XbceXfgoXoF p p p p p FFHHHHHrctj|r'|tj|jjdz}|tj|jdz}|jjd}tjtj|j|kd}|j||_| |t|z S)a  clips out all data points that are more than a certain number of standard deviations from the mean. sigs can be either a single value or a length-p sequence that specifies the number of standard deviations along each of the p dimensions. rr r) risscalaronesrr%rallabsrr,rI)rsigsr&r s rsigclipz Pca.sigclipws ;t   /bgdfl1o...DBF46q)))) FLO F26$&>>D(a 0 0 0vay Qxrcj|j|_|dSN)r+r*rr,r8s rresetz Pca.resets(!!## rctd|||fD}|dkrtd}n|dkrtd|||k}nO|td|}n<|+t j||k}nt d| |jj}nKt j |d}|jjj d|j dkrtd t j | j|z}||jS) aD projects the normalized values onto the components enthresh, nPCs, and cumen determine how many PCs to use if vals is None, the normalized data vectors are the values to project. Otherwise, it should be convertable to a p x N array returns n,p(>threshold) dimension array cg|]}|duSrr)r es rr"zPca.project..sGGG! GGGrrNrz&cannot specify more than one thresholdzShould be unreachableF)r*zshape for vals does not match) rIslicer1energiesrcumsum RuntimeErrorrr$r#r%matrixrG)rr_enthreshnPCscumennonnonesr projs rprojectz Pca.projects7GG$/FGGGHH q==d AA \\EFF F#MMOOh.!$t$$"Idmmoo..%7"#:;;; <68DD8De,,,Dvx~a DJqM11 !@AAAy--//00247Awyrcftj|}|j\}}|jjd}||krt dtjtj|j }tj ||f}||ddd|f<||j z}|rtj |j j Stj |jd} tj |jd} tj |j | z| zj S)zP input is an n X q array, where q <= p output is p X n rzq > pNrr )r atleast_2dr%rr1r<invrrGr$rsr#rrr) rrnormedr&qr'evinvzsrmnssdss r deprojectz Pca.deprojects -  g! FLO q55W%% %immBId&:&:&<&<==?@@ Xqe__111RaR4RTz  08DF##% %A&&&Ctv1%%%CHTV$$S(,/ /rc||j}n7|j}|jd|jjdkrtd|}t j|}|dd|f|dd|f<||d}|j|z }|jt j|j dz}|t j |jdzS)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$r%r1rr zeros_likerrrr)rpcr_pcszpcsupcrBs r subtractPCzPca.subtractPCs <6DD6Dz!}tv|A.. !TUUU LLNN ]3  qqqt9QQQrT NN4 & & F3J Ctv1%%% %Q'''''rr)rrT)rrrjT)NNNN)T)__name__ __module__ __qualname____doc__r.r,r5r9rArDrGrKriryrrrrrrrrrrs *GOOOO*     ((((((77774($ !!!!F00004((((((rr)numpyrrrrrrsM[([([([([([([([([([(r