U f{j@sJdZddlZddlmZddlmZddlZddlZddlZddZ ddZ d\d d d d Z ddZ ddZ ddZd]ddZddZddZddZddZddZd d!Zd"d#Zd d d d$gd d d d d d d d d d d%ejgd d d&ZGd'd(d(eZd)d*Zd^d+d,Zd_d-d.ZGd/d0d0eZd1d2Zd3d4Z d5d6Z!d7d8Z"d`d9d:Z#dad=d>Z$dbd?d@Z%dAdBZ&z0ddl'Z'e e'j(dCdDre)e'j*Z*dcdEdFZ+Wne)k rd Z,YnXe-dGdH.dHkZ,e,sdIdFZ+e/Z*ej0Z0ej1Z1n*e+dJdKZ0e+dLdMZ2e+dNdOZ1dPdQZ3dRdSZ4dddTdUZ5dedVdWZ6dXdYZ7dZd[Z8dS)fzCompatibility fixes for older versions of libraries If you add content to this file, please give the version of the package at which the fix is no longer needed. # originally copied from scikit-learn N)log)pprintc CsVddlm}tjdd4tdtd|d|d|d W5QRSQRXd S) aCompare two version strings via a user-specified operator. Parameters ---------- version_a : str First version string. operator : '==' | '>' | '<' | '>=' | '<=' Operator to compare ``version_a`` and ``version_b`` in the form of ``version_a operator version_b``. version_b : str Second version string. Returns ------- bool The result of the version comparison. r)parseTrecordignorezparse("z") z parse("z")N)Zpackaging.versionrwarningscatch_warnings simplefiltereval)Z version_aoperatorZ version_brr H/var/www/html/bwave/python/venv/lib/python3.8/site-packages/mne/fixes.py_compare_versions  rcCsJt|r8tjt||ddtjt||d}ntj||d}|S)zCompute marginal median on complex data safely. Can be removed when numpy introduces a fix. See: https://github.com/scipy/scipy/pull/12676/. axisy?)np iscomplexobjZmedianrealimag)datarr r r_median_complex:s  rF)Zvar_positionalZ var_keywordexcludecsjt|j}tddDfdd|D}|rbdd|D}t|dkrZd}||fS|SdS)Ncss|]}ttj|VqdSN)getattrinspect Parameterupper).0exr r r Qsz_get_args..csg|]\}}|jkr|qSr )kind)rkeyparamrr r Rs z_get_args..cSsg|]}|j|jkr|jqSr )r"VAR_POSITIONALname)rr$r r rr%Ts r)r signature parameterssetitemsvalueslen)functionvarargsrparamsargsr rr _get_argsJs  r2c Ksddlm}|ddr tdz|j|f|WStjjk r}z8ddlm}|d|f|j|fd d i|WYSd }~XYnXd S) z=Wrapper to get around the SVD did not converge error of deathrlinalgZ overwrite_aFz.Cannot set overwrite_a=True with this function)warnz8SVD error (%s), attempting to use GESVD instead of GESDDZ lapack_driverZgesvdN) scipyr4get ValueErrorsvdrZ LinAlgErrorutilsr6)Akwargsr4expr6r r r _safe_svd]s   r?cCsddlm}||S)Nr) csc_matrix)Z scipy.sparser@)xr@r r r_csc_matrix_castos rBcCsDzddl}d}Wntk r(d}YnX|ri2gY@)>f4?dtyper utf-8>i4rJz/File does not appear to be a Freesurfer surfacez+No volume information contained in the file)ZsurfacerGrHdictopenrfromfileastypefloat64reshapezerosZint64readlinerstripdecode_read_volume_infor9r-rr6)filepathZ read_metadataZ read_stamprGrH volume_infoZTRIANGLE_MAGICZ QUAD_MAGICZNEW_QUAD_MAGICfobjmagicZnvertZnquadfmtdivZcoordsZquadsZfacesZnfaceZquadZ create_stampZvnumZfnumretr r rrEs\          rEcCst|dt|ddS)z$Get the unform/randint from the rng.ZintegersrandintNr)rngr r r rng_uniformsrjcCs^t|}|jdkrtd|j\}}|dkr6td|dddfdksVtd||fS) zHelper to validate a SOS inputrOzsos array must be 2Dz'sos array must be shape (n_sections, 6)NrLr5zsos[:, 3] should be all ones)rZ atleast_2dndimr9shapeall)sosZ n_sectionsmr r r _validate_soss   rqcCs,t|j}t|rtjntj}||Sr)rZ asanyarrayZdataobjrZ complex128rYrX)ZimgrrQr r r_get_img_fdatas rrcCst}t|dd}t|dgsZt|t|ddg}t|dddgsZtd|S||d<dD]}|d  d }|d |kst |dkrt d |d kr|d ||<qf|d krt |d t||<qft |d t||<qf|S)z}An implementation of nibabel.freesurfer.io._read_volume_info, since old versions of nibabel (<=2.1.0) don't have it. rTr5rOrUnknown extension code.head)validfilenamevolume voxelsizexrasyraszrascrasrS=zError parsing volume info.rvrwrx)rUrrW array_equalZ concatenaterr6r\r^splitstripr-IOErrorarrayrXintfloat)rbrarur#pairr r rr_s$ r_c CsRddddddddd g }t||}t|d krDtd |t}|D]}|dkrt||d gst||d d d gst d| tj ||dd qN|dkr||}| d||dqN|dkr||}| d||d |d|d dqN||}| d|d|d |d|d dqNd|S)z}An implementation of nibabel.freesurfer.io._serialize_volume_info, since old versions of nibabel (<=2.1.0) don't have it.rurvrwrxryrzr{r|r}rzInvalid volume info: %s.rsrOrtrTrPrz{} = {} rSz{} = {} {} {} r5z {} = {:0.10g} {:0.10g} {:0.10g} rk)r*keys differencer-r9poplistrrrr6appendrtobytesformatencodeljustjoin)rardiffstringsr#valr r r_serialize_volume_infosN   rcCst|dddkS)aReturns True if the given estimator is (probably) a classifier. Parameters ---------- estimator : object Estimator object to test. Returns ------- out : bool True if estimator is a classifier and False otherwise. _estimator_typeN classifierrhZ estimatorr r r is_classifier"s rcCst|dddkS)aReturns True if the given estimator is (probably) a regressor. Parameters ---------- estimator : object Estimator object to test. Returns ------- out : bool True if estimator is a regressor and False otherwise. rNZ regressorrhrr r r is_regressor2s rZ2darrayT)Znon_deterministicZrequires_positive_XZrequires_positive_yZX_typesZ poor_scoreZ no_validationZ multioutput allow_nanZ statelessZ multilabelZ _skip_testZ _xfail_checksZmultioutput_only binary_onlyZ requires_fitZpreserves_dtypeZ requires_ypairwisec@sFeZdZdZeddZdddZddZd d Zd d Z d dZ dS) BaseEstimatorzBase class for all estimators in scikit-learn. Notes ----- All estimators should specify all the parameters that can be set at the class level in their ``__init__`` as explicit keyword arguments (no ``*args`` or ``**kwargs``). cCstt|jd|j}|tjkrgSt|}dd|jD}|D] }|j|jkr@t d||fq@t dd|DS)z%Get parameter names for the estimatorZdeprecated_originalcSs&g|]}|jdkr|j|jkr|qSself)r'r" VAR_KEYWORDrpr r rr%ps z2BaseEstimator._get_param_names..zscikit-learn estimators should always specify their parameters in the signature of their __init__ (no varargs). %s with constructor %s doesn't follow this convention.cSsg|] }|jqSr )r'rr r rr%{s) r__init__objectrr(r)r,r"r& RuntimeErrorsorted)clsinitZinit_signaturer)rr r r_get_param_namesbs   zBaseEstimator._get_param_namesTc st}|D]tdtzDtjdd}t|d}W5QRXt |r`|dj tkr`WqW5tjdX|rt |dr| }|fdd|D||<q|S) aXGet parameters for this estimator. Parameters ---------- deep : bool, optional If True, will return the parameters for this estimator and contained subobjects that are estimators. Returns ------- params : dict Parameter names mapped to their values. alwaysrTrN get_paramsc3s"|]\}}d||fVqdS)__Nr )rkrr#r rr!sz+BaseEstimator.get_params..)rUrrr DeprecationWarningfiltersrr rr-categoryhasattrrr+update)rdeepoutwvalueZ deep_itemsr rrr}s     zBaseEstimator.get_paramsc Ks|s|S|jdd}|D]\}}|dd}t|dkrv|\}}||kr\td||f||}|jf||iq||krtd||jjft|||q|S)aSet the parameters of this estimator. The method works on simple estimators as well as on nested objects (such as pipelines). The latter have parameters of the form ``__`` so that it's possible to update each component of a nested object. Parameters ---------- **params : dict Parameters. Returns ------- inst : instance The object. Trrr5zsInvalid parameter %s for estimator %s. Check the list of available parameters with `estimator.get_params().keys()`.) rr+rr-r9 set_params __class____name__setattr) rr0Z valid_paramsr#rrr'Zsub_nameZ sub_objectr r rrs&    zBaseEstimator.set_paramscCs |jj}d|t|jddfS)Nz%s(%s)Fr)rrrr)r class_namer r r__repr__szBaseEstimator.__repr__cCstSr) _DEFAULT_TAGSrr r r _more_tagsszBaseEstimator._more_tagscCs<i}tt|jD]"}t|dr||}||q|S)Nr)reversedrgetmrorrrr)rZcollected_tagsZ base_classZ more_tagsr r r _get_tagss    zBaseEstimator._get_tagsN)T) r __module__ __qualname____doc__ classmethodrrrrrrr r r rrXs   %,rcCs6zddlm}Wn tk r0ddlm}YnX|S)Nr check_scoring)Zsklearn.metricsrrDZsklearn.metrics.scorerrr r r_get_check_scorings rcs^zddlm}||WStk rXddlmfdd|D}|YSXdS)aCheck and validate the parameters passed during `fit`. Parameters ---------- X : array-like of shape (n_samples, n_features) Data array. fit_params : dict Dictionary containing the parameters passed at fit. indices : array-like of shape (n_samples,), default=None Indices to be selected if the parameter has the same size as `X`. Returns ------- fit_params_validated : dict Validated parameters. We ensure that the values support indexing. r)_check_fit_params) _validationcs i|]\}}||qSr )Z_index_param_value)rrvXrindicesr r sz%_check_fit_params..N)Zsklearn.utils.validationrrDZsklearn.model_selectionrr+)rZ fit_paramsrZ_sklearn_check_fit_paramsZfit_params_validatedr rrrs  rcCst|}|jdkr t|d}|jddkr8td|rVt|j||jd}ntj |jdd}|jdkr~t |gg}|S)aBComputes the Maximum likelihood covariance estimator Parameters ---------- X : ndarray, shape (n_samples, n_features) Data from which to compute the covariance estimate assume_centered : Boolean If True, data are not centered before computation. Useful when working with data whose mean is almost, but not exactly zero. If False, data are centered before computation. Returns ------- covariance : 2D ndarray, shape (n_features, n_features) Empirical covariance (Maximum Likelihood Estimator). r5)r5rMrzBOnly one sample available. You may want to reshape your data array)Zbias) rZasarrayrlrZrmrr6dotTZcovr)rassume_centered covariancer r rempirical_covariances     rc@sPeZdZdZdddZddZdd Zdd d Zdd dZdddZ ddZ d S)EmpiricalCovarianceaMaximum likelihood covariance estimator Read more in the :ref:`User Guide `. Parameters ---------- store_precision : bool Specifies if the estimated precision is stored. assume_centered : bool If True, data are not centered before computation. Useful when working with data whose mean is almost, but not exactly zero. If False (default), data are centered before computation. Attributes ---------- covariance_ : 2D ndarray, shape (n_features, n_features) Estimated covariance matrix precision_ : 2D ndarray, shape (n_features, n_features) Estimated pseudo-inverse matrix. (stored only if store_precision is True) TFcCs||_||_dSr)store_precisionr)rrrr r rrUszEmpiricalCovariance.__init__cCs0ddlm}||_|jr&|||_nd|_dS)arSaves the covariance and precision estimates Storage is done accordingly to `self.store_precision`. Precision stored only if invertible. Parameters ---------- covariance : 2D ndarray, shape (n_features, n_features) Estimated covariance matrix to be stored, and from which precision is computed. rr3N)r7r4 covariance_rpinvh precision_)rrr4r r r_set_covarianceYs z#EmpiricalCovariance._set_covariancecCs*ddlm}|jr|j}n ||j}|S)zGetter for the precision matrix. Returns ------- precision_ : array-like, The precision matrix associated to the current covariance object. rr3)r7r4rrrr)rr4 precisionr r r get_precisionps  z!EmpiricalCovariance.get_precisionNcCsB|jrt|jd|_n |d|_t||jd}|||S)aFit the Maximum Likelihood Estimator covariance model. Parameters ---------- X : array-like, shape = [n_samples, n_features] Training data, where n_samples is the number of samples and n_features is the number of features. y : ndarray | None Not used, present for API consistency. Returns ------- self : object Returns self. r5rr)rrr[rm location_meanrr)rryrr r rfits  zEmpiricalCovariance.fitcCs$t||jdd}t||}|S)aCompute the log-likelihood of a Gaussian dataset. Uses ``self.covariance_`` as an estimator of its covariance matrix. Parameters ---------- X_test : array-like, shape = [n_samples, n_features] Test data of which we compute the likelihood, where n_samples is the number of samples and n_features is the number of features. X_test is assumed to be drawn from the same distribution than the data used in fit (including centering). y : ndarray | None Not used, present for API consistency. Returns ------- res : float The likelihood of the data set with `self.covariance_` as an estimator of its covariance matrix. Tr)rrlog_likelihoodr)rZX_testrZtest_covresr r rscores zEmpiricalCovariance.score frobeniusc Csddlm}||j}|dkr.t|d}n,|dkrRt|t|j|}nt d|rl||j d}|rv|}n t |}|S)a Computes the Mean Squared Error between two covariance estimators. Parameters ---------- comp_cov : array-like, shape = [n_features, n_features] The covariance to compare with. norm : str The type of norm used to compute the error. Available error types: - 'frobenius' (default): sqrt(tr(A^t.A)) - 'spectral': sqrt(max(eigenvalues(A^t.A)) where A is the error ``(comp_cov - self.covariance_)``. scaling : bool If True (default), the squared error norm is divided by n_features. If False, the squared error norm is not rescaled. squared : bool Whether to compute the squared error norm or the error norm. If True (default), the squared error norm is returned. If False, the error norm is returned. Returns ------- The Mean Squared Error (in the sense of the Frobenius norm) between `self` and `comp_cov` covariance estimators. rr3rrOZspectralz1Only spectral and frobenius norms are implemented) r7r4rrsumZamaxZsvdvalsrrNotImplementedErrorrmsqrt) rZcomp_covnormZscalingZsquaredr4errorZ squared_normresultr r r error_norms   zEmpiricalCovariance.error_normcCs.|}||j}tt|||d}|S)aComputes the squared Mahalanobis distances of given observations. Parameters ---------- observations : array-like, shape = [n_observations, n_features] The observations, the Mahalanobis distances of the which we compute. Observations are assumed to be drawn from the same distribution than the data used in fit. Returns ------- mahalanobis_distance : array, shape = [n_observations,] Squared Mahalanobis distances of the observations. r5)rrrrr)rZ observationsrZ centered_obsZmahalanobis_distr r r mahalanobiss zEmpiricalCovariance.mahalanobis)TF)N)N)rTT) rrrrrrrrrrrr r r rr;s    0rcCsF|jd}t|| t|}||tdtj8}|d}|S)a9Computes the sample mean of the log_likelihood under a covariance model computes the empirical expected log-likelihood (accounting for the normalization terms and scaling), allowing for universal comparison (beyond this software package) Parameters ---------- emp_cov : 2D ndarray (n_features, n_features) Maximum Likelihood Estimator of covariance precision : 2D ndarray (n_features, n_features) The precision matrix of the covariance model to be tested Returns ------- sample mean of the log-likelihood rrO@)rmrr_logdetrpi)Zemp_covrrZlog_likelihood_r r rrs  rcCsTddlm}||}||jttjj}t ||k||}t t |S)z6Compute the log det of a positive semidefinite matrix.rr3) r7r4ZeigvalshmaxsizerfinforYepswhererr)r<r4valsZtolr r rrs   rcCs:t|}t|}t|D]}t||||||<q|S)zInfers the dimension of a dataset of shape (n_samples, n_features) The dataset is described by its spectrum `spectrum`. )r-remptyrange_assess_dimension_argmax)spectrum n_samples n_featuresZ n_spectrumllrankr r r_infer_dimension_'s   rc Csddlm}|t|kr td| td}t|D].}||||dttj||d7}q6tt|d|}| |d}||krd}d} n6t||d||} t|  |||d}||||dd} tdtj| |dd} d} | } | | ||<t|D]X}t|dt|D]>}| t||||d| |d| |t|7} q@q*|||| | d|t|d}|S)Nr)gammalnz5The tested rank cannot exceed the rank of the datasetrr5rKg) Z scipy.specialrr-r9rrrrrcopy)rrrrrZpuiplpvrrppppaZ spectrum_jrr r rr2s:     (rcCs|rVtjt|dd}t||t|jdf}||9}||ddtjf9}nPtjt|dd}t|t|jd|f}||9}||ddtjf9}||fS)Nrrr5)rrabssignZarangermnewaxis)urZu_based_decisionZ max_abs_colsZsignsZ max_abs_rowsr r rsvd_flipWsrh㈵>:0yE>c CsXtj||tjd}tj||tjd}ttj|jd|d|||ddsTtdt |S)aUse high precision for cumsum and check that final value matches sum Parameters ---------- arr : array-like To be cumulatively summed as flat axis : int, optional Axis along which the cumulative sum is computed. The default (None) is to compute the cumsum over the flattened array. rtol : float Relative tolerance, see ``np.allclose`` atol : float Absolute tolerance, see ``np.allclose`` )rrQrMrT)rtolatolZ equal_nanzLcumsum was found to be unstable: its last element does not correspond to sum) rZcumsumrYrrniscloseZtakerr6RuntimeWarning)arrrrrrexpectedr r r stable_cumsumgsrcCs|rtj|\}}t|}t|}t|ddddf}tj||dd}tj||dd}tj||ddddfdd}||dtjddfdd }tj||d|||fStj |SdS)N.rMr)r) rr4eighrr ZargsortZtake_along_axisrZswapaxesconjr:)aZ hermitiansrZsgnZsidxvtr r rr:s  $ r:c Csddl}|dkr|dkrdS|jj}|dkr4|}|dkrD|}t||t|}t|j ddr>|j ||| d}t |d|dg}t ||g||gg}|jd}dd|d|dd|dd|d|d|dg } |jd| } |jd| } |jdkr"t| | gn t| | g} |j| n|j |||||j} | dddf||7<| dd dfd ||8<| d ddf||7<|j| |||dS) zi crop a colorbar to show from cbar_vmin to cbar_vmax Used when symmetric_cbar=False is used. rNz>=z3.2.0rMr5rO horizontalrkrK) matplotlibZlocatorZlocsrminrZlinspacer-r __version__axZset_ylimZ_meshrrmrrZZ orientationZ column_stackoutlineZset_xyrZget_xyZ set_ticksZ update_ticks)ZcbarZ cbar_vminZ cbar_vmaxr"Zcbar_tick_locsZ new_tick_locsrYNiirArZxyr&r r r_crop_colorbarsB  2  "  r*._jitr )r=r4r r rr1scCs2t|}t||D]\}}|||7<q|Sr)rr[zip)rAweightsZ minlengthridxrr r rbincounts r8cCs|jdkst|dkst|dkr^t|jd}tt|D]}||dd|f||<q>n:t|jd}tt|D]}|||ddf||<qz|S)NrO)rr5rr5)rlAssertionErrorrrrmrr-)Zfunc1drrrrr r r_np_apply_along_axiss r:cCsttj||Sr)r:rr)rrr r rrsrcCsXddlm}z |j|d|dWn&tk rR|d|dYnXdS)z,Force calling of the MPL figure close event.r5)loggerzCalled z.canvas.close_event()zCalling z.canvas.close_event() failedN)r;r;ZcanvasZ close_eventdebugr9)Zfigr;r r r _close_events  r=cCs8z|WStk r*|YSX|Sr)Zget_subplotspecZ is_last_rowAttributeError)r%r r r _is_last_row s r?cCs|tj|\}}~|dkr.|jt|jj}tt|}t|||k}d||}|dd|f}||| j S)z/Compute a pseudo-inverse of a Hermitian matrix.NrK) rr4rrrrQrrr rr)rrrrmaxSZ above_cutoffZ psigma_diagr r rrs rcCstjj|dd\}}}~t|}|dkrHt|j|jt|jj}t|||k}|ddd|f}||d|}||d| j S)z%Compute a pseudo-inverse of a matrix.F)Z full_matricesN) rr4r:rrmrrQrrrr)rrrrZvhr@rr r rpinv%s rAcCs(z|jWStk r"|jYSXdSr)Z point_datar>Z point_arraysobjr r r _point_data8srDcCs(z|jWStk r"|jYSXdSr)Z cell_datar>Z cell_arraysrBr r r _cell_data?srE)F)FF)N)F)T)Nrr)F)TTTT)N)N)9rrmathrrosrnumpyrrrr2r?rBrFrErjrqrrr_rrrrYrrrrrrrrrrrrrr:r*r0r$rDZpranger1Z has_numbagetenvlowerrr8rr:r=r?rrArDrEr r r rs     A" % 'G  %   2