a h2@sdZddlZddlmZddlZddlmZddl m Z m Z m Z ddl mZddlmZmZdd lmZdd lmZdd lmZdd lmZmZdd lmZddlmZddlm Z m!Z!Gdddee e Z"dS)z! Nearest Centroid Classification N)Real)sparse) BaseEstimatorClassifierMixin _fit_context)#DiscriminantAnalysisPredictionMixin)pairwise_distancespairwise_distances_argmin) LabelEncoder)get_tags) available_if)Interval StrOptions)check_classification_targets)csc_median_axis_0)check_is_fitted validate_datacseZdZUdZeddhgeedddddgded d hgd Zee d <ddd d ddZ e ddddZ fddZ ddZddZeeejZeeejZeeejZfddZZS)NearestCentroida Nearest centroid classifier. Each class is represented by its centroid, with test samples classified to the class with the nearest centroid. Read more in the :ref:`User Guide `. Parameters ---------- metric : {"euclidean", "manhattan"}, default="euclidean" Metric to use for distance computation. If `metric="euclidean"`, the centroid for the samples corresponding to each class is the arithmetic mean, which minimizes the sum of squared L1 distances. If `metric="manhattan"`, the centroid is the feature-wise median, which minimizes the sum of L1 distances. .. versionchanged:: 1.5 All metrics but `"euclidean"` and `"manhattan"` were deprecated and now raise an error. .. versionchanged:: 0.19 `metric='precomputed'` was deprecated and now raises an error shrink_threshold : float, default=None Threshold for shrinking centroids to remove features. priors : {"uniform", "empirical"} or array-like of shape (n_classes,), default="uniform" The class prior probabilities. By default, the class proportions are inferred from the training data. .. versionadded:: 1.6 Attributes ---------- centroids_ : array-like of shape (n_classes, n_features) Centroid of each class. classes_ : array of shape (n_classes,) The unique classes labels. n_features_in_ : int Number of features seen during :term:`fit`. .. versionadded:: 0.24 feature_names_in_ : ndarray of shape (`n_features_in_`,) Names of features seen during :term:`fit`. Defined only when `X` has feature names that are all strings. .. versionadded:: 1.0 deviations_ : ndarray of shape (n_classes, n_features) Deviations (or shrinkages) of the centroids of each class from the overall centroid. Equal to eq. (18.4) if `shrink_threshold=None`, else (18.5) p. 653 of [2]. Can be used to identify features used for classification. .. versionadded:: 1.6 within_class_std_dev_ : ndarray of shape (n_features,) Pooled or within-class standard deviation of input data. .. versionadded:: 1.6 class_prior_ : ndarray of shape (n_classes,) The class prior probabilities. .. versionadded:: 1.6 See Also -------- KNeighborsClassifier : Nearest neighbors classifier. Notes ----- When used for text classification with tf-idf vectors, this classifier is also known as the Rocchio classifier. References ---------- [1] Tibshirani, R., Hastie, T., Narasimhan, B., & Chu, G. (2002). Diagnosis of multiple cancer types by shrunken centroids of gene expression. Proceedings of the National Academy of Sciences of the United States of America, 99(10), 6567-6572. The National Academy of Sciences. [2] Hastie, T., Tibshirani, R., Friedman, J. (2009). The Elements of Statistical Learning Data Mining, Inference, and Prediction. 2nd Edition. New York, Springer. Examples -------- >>> from sklearn.neighbors import NearestCentroid >>> import numpy as np >>> X = np.array([[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]]) >>> y = np.array([1, 1, 1, 2, 2, 2]) >>> clf = NearestCentroid() >>> clf.fit(X, y) NearestCentroid() >>> print(clf.predict([[-0.8, -1]])) [1] manhattan euclideanrNZneither)closedz array-like empiricaluniformmetricshrink_thresholdpriors_parameter_constraints)rrcCs||_||_||_dS)Nr)selfrrrr _/var/www/html/assistant/venv/lib/python3.9/site-packages/sklearn/neighbors/_nearest_centroid.py__init__szNearestCentroid.__init__T)Zprefer_skip_nested_validationcCs|jdkr"t|||dgd\}}n.t|jjr2dnd}t||||ddgd\}}t|}t||j\}}t }| |}|j |_ } | j } | dkrt d | |jd krtj|dd \} } t| tt||_n0|jd krtd | g| |_nt|j|_|jdkr t dt|jdsRtdt|j|j|_tj| |ftjd|_t| } t | D]}||k}t|| |<|rt!|d}|jdkr|stj"||dd|j|<nt#|||j|<n||j$dd|j|<qztj%||j|ddd}tj%t&|jdd|| dd|_'t|j'dkr^tdd}|rt(|j)dd|j*dd+dkrt |n(|st(tj,|dddkrt ||j$dd}t&d| d|}|j't"|j'}|-t|d }||}tj%|j||dd|_.|j/rt0|j.}t1|j.|j/|_.tj2|j.dd|j.d|j.|9_.||j.}tj%||dd|_|S)a0 Fit the NearestCentroid model according to the given training data. 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. Note that centroid shrinking cannot be used with sparse matrices. y : array-like of shape (n_samples,) Target values. Returns ------- self : object Fitted estimator. rZcsc) accept_sparse allow-nanTcsr)ensure_all_finiter#rz>The number of classes has to be greater than one; got %d classr)Zreturn_inverserrzpriors must be non-negativeg?zAThe priors do not sum to 1. Normalizing such that it sums to one.dtype)ZaxisF)copyzself.within_class_std_dev_ has at least 1 zero standard deviation.Inputs within the same classes for at least 1 feature are identical.z2All features have zero variance. Division by zero.N)out)3rrr input_tags allow_nanspissparsershaper Z fit_transformclasses_size ValueErrorrnpuniqueZbincountfloatlen class_prior_Zasarrayanyisclosesumwarningswarn UserWarningemptyfloat64 centroids_ZzerosrangewhereZmedianrmeanarraysqrtwithin_class_std_dev_allmaxminZtoarrayZptpZreshapeZ deviations_rsignabsZclip)rXyr&Z is_X_sparseZ n_samplesZ n_featuresleZy_indclassesZ n_classes_Z class_countsZnkZ cur_classZ center_maskZvarianceerr_msgZdataset_centroid_msmmmsZsignsZmsdr r r!fits           .     zNearestCentroid.fitcsrt|t|jdt|jrbt|jj r4dnd}t |||ddd}|jt ||j |j dSt|SdS) aPerform classification on an array of test vectors `X`. The predicted class `C` for each sample in `X` is returned. Parameters ---------- X : {array-like, sparse matrix} of shape (n_samples, n_features) Input data. Returns ------- y_pred : ndarray of shape (n_samples,) The predicted classes. r'r$Tr%F)r&r#resetrN)rr4r:r8r7r1rHr r,r-rr rArsuperpredict)rrMr& __class__r r!r[szNearestCentroid.predictc Cst|dt||dddtjd}tj|jd|jjftjd}|jdk}|dd|f|j|<|j }|dd|f|j|<t |jjD]R}t |||g|j d}|d C}t| d t|j||dd|f<q|S) NrATFr%)r*rXr#r)rr(rYrg@)rrr4r@r?r0r1r2rGrAr*rBr rZravelZsqueezelogr8)rrMZ X_normalizedZdiscriminant_scoremaskZcentroids_normalizedZ class_idxZ distancesr r r!_decision_function8s(    z"NearestCentroid._decision_functioncCs |jdkS)NrrY)rr r r!_check_euclidean_metricTsz'NearestCentroid._check_euclidean_metriccs$t}|jdk|j_d|j_|S)NZ nan_euclideanT)rZ__sklearn_tags__rr,r-r)rtagsr\r r!rbcs z NearestCentroid.__sklearn_tags__)r)__name__ __module__ __qualname____doc__rrrrdict__annotations__r"rrWr[r`rar rZdecision_functionZ predict_probaZpredict_log_probarb __classcell__r r r\r!rs2 h   } "r)#rgr<numbersrnumpyr4Zscipyrr.baserrrZdiscriminant_analysisrZmetrics.pairwiser r Z preprocessingr utilsr Zutils._available_ifr Zutils._param_validationrrZutils.multiclassrZutils.sparsefuncsrZutils.validationrrrr r r r!s"