:jddlZddlmZmZddlmZmZddlZddl m Z ddl m Z mZmZddlmZmZddlmZddlmZddlmZdd lmZdd lmZmZmZm Z m!Z!dd l"m#Z#m$Z$m%Z%dd l&m'Z'dd l(m)Z)ddl*m+Z+ddl,m-Z-ddl.m/Z/m0Z0ddl1m2Z2m3Z3m4Z4m5Z5m6Z6m7Z7gdZ8dZ9Gdde eZ:Gddee:eZ;dZ< ddZ=y)N)ABCMetaabstractmethod)IntegralReal) BaseEstimatorClassifierMixin _fit_context)ConvergenceWarningNotFittedError) LabelEncoder) _liblinear)_libsvm)_libsvm_sparse) check_arraycheck_random_state column_or_1dcompute_class_weight deprecated)HiddenInterval StrOptions)_align_api_if_sparse)safe_sparse_dot)SCIPY_VERSION_BELOW_1_12) available_if)_ovr_decision_functioncheck_classification_targets)_check_large_sparse_check_sample_weight _num_samplescheck_consistent_lengthcheck_is_fitted validate_data)c_svcnu_svc one_class epsilon_svrnu_svrc |jddz}g}tjtjdg|g}t |D]}|||||dzddf}t |dz|D]}|||||dzddf} t r*||dz g||||dzf} ||g||||dzf} n'||dz ||||dzf} ||||||dzf} |j t| |t| | z|S)zGenerate primal coefficients from dual coefficients for the one-vs-one multi class LibSVM in the case of a linear kernel.rN)shapenpcumsumhstackrangerappendr) dual_coef n_supportsupport_vectorsn_classcoefsv_locsclass1sv1class2sv2alpha1alpha2s f/home/mcse/projects/srt_converter/srt-converter-venv/lib/python3.12/site-packages/sklearn/svm/_base.py_one_vs_one_coefr>0soooa 1$G Dii A3 "234G.Ugfo 0CCQFGFQJ0 UF!'&/GFQJ4G"G"JKC'"FQJ<76TU:CV1V#VW"F8WV_wvPQz?R-R#RS#6A:wvRSAT/T#TU"676?WVaZ=P+P#PQ KK4vs7SS T! UU( KceZdZUdZehdegeedddgeddheed ddgeeddd geed dd geed dd geed d d geed ddgd gd e edhgeeddd gedhe dgdgeedddgdgdZ e e d<gdZ edZfdZedd*dZdZdZdZdZdZd Zd!Zd"Zd#Zd$Zd%Zd&Zed'Z d(Z!ed)Z"xZ#S)+ BaseLibSVMzBase class for estimators that use libsvm as backing library. This implements support vector machine classification and regression. Parameter documentation is in the derived `SVC` class. >rbfpolylinearsigmoid precomputedrNleft)closedscaleautoneitherright?booleanrbalancedverbose random_statekerneldegreegammacoef0tolCnuepsilon shrinking probability cache_size class_weightrQmax_iterrS_parameter_constraints)rDrCrBrErFc:|jtvr tdtd|jd||_||_||_||_||_||_||_ ||_ | |_ | |_ | |_ | |_| |_||_||_y)Nzimpl should be one of z, z was given)_impl LIBSVM_IMPL ValueErrorrUrVrWrXrYrZr[r\r]r^r_r`rQrarS)selfrUrVrWrXrYrZr[r\r]r^r_r`rQrarSs r=__init__zBaseLibSVM.__init__{s& ::[ (@" /4+!O !  QWWQZ 757@!''!*MN  ;;- 'I,C,,2F1771:qwwqz,J    q !A %-*=*=a*@I*M !&& 1 #+4;;"7T[[ ] "DK  C (zzW$DJA,,.!&&(q@PQPUPUPW>DL  >>..0DL r?cZt|djtjdS)zxValidation of y and class_weight. Default implementation for SVR and one-class; overridden in BaseSVC. TrF)r)rastyper,r)rgrs r=rzBaseLibSVM._validate_targetsJs% AD)00%0HHr?c|jdvsJ|jdk(r(tjd|jztyy)Nrr*r*znSolver terminated early (max_iter=%i). Consider pre-processing your data with StandardScaler or MinMaxScaler.) fit_status_rrrar rgs r=_warn_from_fit_statusz BaseLibSVM._warn_from_fit_statusQsL6)))   q MM359]]C#   !r?c  t|jrB||_|j|}|jd|jdk7r t dt j|jt j||fid|d|dt|dtjdd|d |jd |jd |jd |j d |j"d|j$d|j&d|j(d|j*d|j,d|j.d|\ |_|_|_|_|_|_|_|_|_ |jCy)Nrr*z(X.shape[0] should be equal to X.shape[1]svm_typerr` class_weight_rUrZr[r^rVr]rYr_rXrWr\rar)"rrU_BaseLibSVM__Xfit_compute_kernelr+rflibsvmset_verbosity_wraprQrgetattrr,emptyrZr[rrVr]rYr_rXrr\rasupport_support_vectors_ _n_supportrr_probA_probBrrr)rgrrrrrUrs r=rzBaseLibSVM._dense_fit[s DKK DK$$Q'AwwqzQWWQZ' !KLL!!$,,/ JJ  ! (  ! D    ff ww 33 ;; nn   ** ++ LL! "]]# $$%  M  ! O O O K K   N, ""$r?c>tj|jtjd|_|j |j j |}tj|jtj|jd|j|j|j||||j|j|j |j"|j$t'|dtj(d||j*|j,|j.t1|j2t1|j4|j6|\ |_|_}|_|_|_ |_!|_"|_#|jItK|drtM|jNdz } nd} |j:jd} tjPtjR| | } | s%tUtWjXgg|_-ytjRd| j\dz| j\| z } tUtWjX|| | f| | f|_-y)NrZrvrwr*rrr)/r,rrr sort_indices_sparse_kernelsr libsvm_sparserrQlibsvm_sparse_trainr+indicesindptrrVrrXrYrZrrr[r_r\intr]rrarrrrrrrrrrrrtilearangerr csr_arrayrsize) rgrrrrrUr kernel_typedual_coef_datar4n_SVdual_coef_indicesdual_coef_indptrs r=rzBaseLibSVM._sparse_fitsAFF"**C@ **008 ((6  - - GGAJ FF II HH   KK KK JJ HH FF D/288A; 7  GG OO LL   ++ , MM +  M  !  O O K K   N2 ""$ 4 $$--(1,GG$$**1-GGBIIdOW=22<<3EFDO!yy$))A-/@/E/E/O  3 #%68HIdODOr?c||j|}|jr |jn |j}||S)aPerform regression on samples in X. For a one-class model, +1 (inlier) or -1 (outlier) is returned. Parameters ---------- X : {array-like, sparse matrix} of shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is (n_samples_test, n_samples_train). Returns ------- y_pred : ndarray of shape (n_samples,) The predicted values. )_validate_for_predictr_sparse_predict_dense_predict)rgrpredicts r=rzBaseLibSVM.predicts7  & &q )*.,,$&&D(a9 NN  ;;q>$,,x 7LEt*%%&  ;;- 'wwqzT__Q// =wwqz4??1#567 " "||! 0C0C0ERS0T1$..2I2I1J,W r?c|jdk7r td|j}tj|rd|j j _|Sd|j _|S)zWeights assigned to the features when `kernel="linear"`. Returns ------- ndarray of shape (n_features, n_classes) rDz2coef_ is only available when using a linear kernelF)rUAttributeError _get_coefrrrflags writeablergr5s r=coef_zBaseLibSVM.coef_s` ;;( " !UV V~~ ;;t (-DIIOO % $)DJJ  r?cBt|j|jSN)rrrrs r=rzBaseLibSVM._get_coefst//1F1FGGr?c t|tj |j }|dvr |j Stj|j dgS#t$rtwxYw)z)Number of support vectors for each class.rr) r"r rrerrdrr,array)rgrs r=rzBaseLibSVM.n_support_sk ! D !$$TZZ0 v ?? "88T__Q/01 1 !  !s AA/r )$r __module__ __qualname____doc__rrrrrrdictrb__annotations__rrrhrlr rrrrrrrrrrrrrpropertyr rr __classcell__rrs@r=rArAVs J K  Haf=> ( ) T3V 4 4tI>?sD;<tS$w7 8c3w78T3V<=[!6*l^*D#EFai@A#ZL14>;hD@A'(+$D6JO%)%)N5]6]~I-%^>@(  D" H < 0! F'R,H 2 2r?rA) metaclassc^eZdZUdZiej eddhgdgdZeed<dD]Z eje e fdZ d Z d Zfd Zd Zeed ZeedZdZdZdZededZededZfdZxZS)BaseSVCz!ABC for LibSVM-based classifiers.ovrovorO)decision_function_shape break_tiesrb)r\r[c^||_||_t| |||||||d|| | | | | |y)NrKrT)rrrkrh)rgrUrVrWrXrYrZr[r]r^r_r`rQrarrSrrrs r=rhzBaseSVC.__init__sS((?$$ #!%%  r?cDt|d}t|tj|d\}}t |j |||_t|dkrtdt|z||_ tj|tjdS) NTr)return_inverse)classesrrz>The number of classes has to be greater than one; got %d classrZr) rrr,uniquerr`rrrfrrr)rgry_clss r=rzBaseSVC._validate_targetss !$ '$Q'2d3Q1$2C2CSTVW s8a<Pc(   zz!2::S99r?c|j|}|jdk(r`_ for further details. If decision_function_shape='ovr', the decision function is a monotonic transformation of ovo decision function. rrr)rrrrr)rgrdecs r=rzBaseSVC.decision_functionsU6%%a(  ' '5 0S5G!5K)#'C4T]]9KL L r?ct||jr|jdk(r td|jrN|jdk(r?t |j dkDr't j|j|d}nt|)|}|j jt j|t jS)aPerform classification on samples in X. For a one-class model, +1 or -1 is returned. Parameters ---------- X : {array-like, sparse matrix} of shape (n_samples, n_features) or (n_samples_test, n_samples_train) For kernel="precomputed", the expected shape of X is (n_samples_test, n_samples_train). Returns ------- y_pred : ndarray of shape (n_samples,) Class labels for samples in X. rz>break_ties must be False when decision_function_shape is 'ovo'rrr*)axisrz)r"rrrfrrr,argmaxrrkrtakerintp)rgrrrrs r=rzBaseSVC.predict>s"  ??t;;uDP  OO,,5DMM"Q& $003!??r?c|jdk(s |js td|jdvr tdy)Nrz5predict_proba is not available when probability=Falser{z0predict_proba only implemented for SVC and NuSVCT)r^rrdrs r= _check_probazBaseSVC._check_probacsF   | +43C3C G  ::0 0 !ST Tr?c|j|}|jjdk(s|jjdk(r t d|j r |j n |j}||S)aCompute probabilities of possible outcomes for samples in X. The model needs to have probability information computed at training time: fit with attribute `probability` set to True. Parameters ---------- X : array-like of shape (n_samples, n_features) For kernel="precomputed", the expected shape of X is (n_samples_test, n_samples_train). Returns ------- T : ndarray of shape (n_samples, n_classes) Returns the probability of the sample for each class in the model. The columns correspond to the classes in sorted order, as they appear in the attribute :term:`classes_`. Notes ----- The probability model is created using cross validation, so the results can be slightly different than those obtained by predict. Also, it will produce meaningless results on very small datasets. rzApredict_proba is not available when fitted with probability=False)rrrrr r_sparse_predict_proba_dense_predict_proba)rgr pred_probas r= predict_probazBaseSVC.predict_probalsq6  & &q ) ;;  q DKK$4$4$9 S +/,,D & &Drrrvstacktocsrr,r s r=rzBaseSVC._get_coefs ??  #q ("4??D4I4IJD $$2G2GD{{47#yy,,. yy r?zAttribute `probA_` was deprecated in version 1.9 and will be removed in 1.11 as the `probability=True` option for SVC and NuSVC was deprecated and will be removed in 1.11.c|jSzParameter learned in Platt scaling when `probability=True`. Returns ------- ndarray of shape (n_classes * (n_classes - 1) / 2) )rrs r=probA_zBaseSVC.probA_{{r?zAttribute `probB_` was deprecated in version 1.9 and will be removed in 1.11 as the `probability=True` option for SVC and NuSVC was deprecated and will be removed in 1.11.c|jSr>)rrs r=probB_zBaseSVC.probB_ r@r?c`t|}|jdk7|j_|Srj)rkrlrUrmrorps r=rlzBaseSVC.__sklearn_tags__s*w')!% != r?)rrrrrArbrrr unused_parampoprrhrrrr-rr2r5r0r/rrrr?rBrlrrs@r=rrs$+$  + +$$.u~$>#? k$D *1 ""<01% % N :@@J," "H,- -:6! F  '    '   r?rc jddiddddddd iidd id d ddddd iiddddidd}|dk(r||S|dk7rtd|z|j|d}|d|z}n@|j|d}| d|d|d}n"|j|d}| d|d|d|}n|Std|d|d|d|)aFind the liblinear magic number for the solver. This number depends on the values of the following attributes: - multi_class - penalty - loss - dual The same number is also internally used by LibLinear to determine which solver to use. Fr)FT)l1l2rJTrr* )logistic_regressionhinge squared_hingeepsilon_insensitivesquared_epsilon_insensitivecrammer_singerrVrz<`multi_class` must be one of `ovr`, `crammer_singer`, got %rNzloss='%s' is not supportedzThe combination of penalty='z ' and loss='z' is not supportedz' are not supported when dual=zUnsupported set of arguments: z, Parameters: penalty=z, loss=z, dual=)rfget) multi_classpenaltylossdual_solver_type_dict _solver_pen error_string _solver_dual solver_nums r=_get_liblinear_solver_typera!s("(-aj8KLq "!& !12EF $tRj1(,b.C'D && --   J[ X  $''d3K3d: "w5  D"  &))$5J!CJ4QUW "!  $ . r?c| dvrVt}|j|}|j}t|dkrt d|dzt ||||}n't jdt j}|}tj|t| }|r tdd d }|r|dkrt d |z|}tj|tj|tj|tj |r t#|t j$|t jj'}t j(|d }t+||t j}t-| || |}tj.||tj ||| |||| |j1t j2dj4|| \}}t5|}|| k\rt7j8dt:|r|ddddf}||dddfz}n|}d}|||fS)aUsed by Logistic Regression (and CV) and LinearSVC/LinearSVR. Preprocessing is done in this function before supplying it to liblinear. Parameters ---------- X : {array-like, sparse matrix} of shape (n_samples, n_features) Training vector, where `n_samples` is the number of samples and `n_features` is the number of features. y : array-like of shape (n_samples,) Target vector relative to X C : float Inverse of cross-validation parameter. The lower the C, the higher the penalization. fit_intercept : bool Whether or not to fit an intercept. If set to True, the feature vector is extended to include an intercept term: ``[x_1, ..., x_n, 1]``, where 1 corresponds to the intercept. If set to False, no intercept will be used in calculations (i.e. data is expected to be already centered). intercept_scaling : float Liblinear internally penalizes the intercept, treating it like any other term in the feature vector. To reduce the impact of the regularization on the intercept, the `intercept_scaling` parameter can be set to a value greater than 1; the higher the value of `intercept_scaling`, the lower the impact of regularization on it. Then, the weights become `[w_x_1, ..., w_x_n, w_intercept*intercept_scaling]`, where `w_x_1, ..., w_x_n` represent the feature weights and the intercept weight is scaled by `intercept_scaling`. This scaling allows the intercept term to have a different regularization behavior compared to the other features. class_weight : dict or 'balanced', default=None Weights associated with classes in the form ``{class_label: weight}``. If not given, all classes are supposed to have weight one. For multi-output problems, a list of dicts can be provided in the same order as the columns of y. The "balanced" mode uses the values of y to automatically adjust weights inversely proportional to class frequencies in the input data as ``n_samples / (n_classes * np.bincount(y))`` penalty : {'l1', 'l2'} The norm of the penalty used in regularization. dual : bool Dual or primal formulation, verbose : int Set verbose to any positive number for verbosity. max_iter : int Number of iterations. tol : float Stopping condition. random_state : int, RandomState instance or None, default=None Controls the pseudo random number generation for shuffling the data. Pass an int for reproducible output across multiple function calls. See :term:`Glossary `. multi_class : {'ovr', 'crammer_singer'}, default='ovr' `ovr` trains n_classes one-vs-rest classifiers, while `crammer_singer` optimizes a joint objective over all classes. While `crammer_singer` is interesting from a theoretical perspective as it is consistent it is seldom used in practice and rarely leads to better accuracy and is more expensive to compute. If `crammer_singer` is chosen, the options loss, penalty and dual will be ignored. loss : {'logistic_regression', 'hinge', 'squared_hinge', 'epsilon_insensitive', 'squared_epsilon_insensitive}, default='logistic_regression' The loss function used to fit the model. epsilon : float, default=0.1 Epsilon parameter in the epsilon-insensitive loss function. Note that the value of this parameter depends on the scale of the target variable y. If unsure, set epsilon=0. sample_weight : array-like of shape (n_samples,), default=None Weights assigned to each sample. Returns ------- coef_ : ndarray of shape (n_features, n_features + 1) The coefficient vector got by minimizing the objective function. intercept_ : float The intercept term added to the vector. n_iter_ : array of int Number of iterations run across for each class. )rTrUrzeThis solver needs samples of at least 2 classes in the data, but the data contains only one class: %rr)r!rrrzz [LibLinear]rrgzqIntercept scaling is %r but needs to be greater than 0. To disable fitting an intercept, set fit_intercept=False.W) requirementsrz@Liblinear failed to converge, increase the number of iterations.NrRrK)r fit_transformrrrfrr,rr liblinearrrrrrrrrrrrequirerra train_wraprrrrrr )rrrZ fit_interceptintercept_scalingr`rYr[rQrarYrSrXrZr\rency_indrrrbiasr raw_coef_r n_iter_maxr rs r=_fit_liblinearrpZs/h IIn!!!$<< x=1 '{+  - (a} "**5    ) \ *C m$ D  !,.?@  %D g&$$W-   ) {{1~A JJuBJJ / 5 5 7E JJu3 /E(LM,['4NK"--   A    BHHSM%%& Iw$WJX N  !SbS&!&1b5)99  *g %%r?)NrrQg?N)>rabcrrnumbersrrnumpyr, scipy.sparseror sklearn.baserrr sklearn.exceptionsr r sklearn.preprocessingr sklearn.svmr rfrrrr sklearn.utilsrrrrrsklearn.utils._param_validationrrrsklearn.utils._sparsersklearn.utils.extmathrsklearn.utils.fixesrsklearn.utils.metaestimatorsrsklearn.utils.multiclassrrsklearn.utils.validationrrr r!r"r#rer>rArrarpr?r=rs'"EEA./*7IH6185H #LC 2'C 2LBozWBJ 6J  !D&r?