ó ‡ˆ\c@s”dZddlmZddlmZddlZddlZddlmZm Z ddl Z ddl m Z mZddlmZdd lmZd d lmZd d lmZd d lmZd dlmZmZmZd dlmZd dlm Z d dl!m"Z"d dl#m$Z$d dl%m&Z&m'Z'm(Z(d dl%m)Z*e+dej,dej,de-ej.ƒƒddddddddd d!d"d#d$d%d&d'd(d)d*d+gƒZ/e+dgdgdej.ƒƒZ0d,„Z1d-„Z2d.e"j3eeƒfd/„ƒYZ4d0„Z5d1e6fd2„ƒYZ7d3„Z8d4e6fd5„ƒYZ9d6e6fd7„ƒYZ:d8e6fd9„ƒYZ;d:e6fd;„ƒYZ<dS(<s,Base and mixin classes for nearest neighborsiÿÿÿÿ(tpartial(t LooseVersionN(tABCMetatabstractmethod(t csr_matrixtissparsei(tBallTree(tKDTreei(t BaseEstimator(tpairwise_distances_chunked(tPAIRWISE_DISTANCE_FUNCTIONS(t check_X_yt check_arraytgen_even_slices(tcheck_classification_targets(tcheck_is_fitted(tsix(tDataConversionWarning(tParalleltdelayedteffective_n_jobs(t __version__t ball_treetkd_treetbrutet braycurtistcanberrat chebyshevt correlationtcosinetdicethammingtjaccardt kulsinskit mahalanobistmatchingt minkowskitrogerstanimotot russellraot seuclideant sokalmichenert sokalsneatht sqeuclideantyulet wminkowskicCs0|dkr|St|ƒr |Stdƒ‚dS(s$Check to make sure weights are validtuniformtdistancesOweights not recognized: should be 'uniform', 'distance', or a callable functionN(NR-R.(tNonetcallablet ValueError(tweights((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt_check_weights3s   cCs|d krdS|dkré|jtjtƒkrx®t|ƒD]H\}}t|dƒr{d|kr{|dk||s"    t NeighborsBasec BsVeZdZed d ddddd d d„ƒZd„Zd„Zed„ƒZ RS( s,Base class for nearest neighbors estimators.tautoiR$ic CsV||_||_||_||_||_||_||_||_|jƒdS(N( t n_neighborstradiust algorithmt leaf_sizetmetrict metric_paramstptn_jobst_check_algorithm_metric( tselfRHRIRJRKRLRNRMRO((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt__init__ms        cCsj|jdkr%td|jƒ‚n|jdkr€|jdkrLd}q‰t|jƒsn|jtdkrwd}q‰d}n |j}t|jƒrÀ|jdkrïtd|jƒ‚qïn/|jt|krïtd|j|fƒ‚n|jdk r3d |jkr3tjd t d d ƒ|jd }n |j }|jdkrf|dkrftdƒ‚ndS(NRGRRRsunrecognized algorithm: '%s't precomputeds7kd_tree algorithm does not support callable metric '%s's†Metric '%s' not valid. Use sorted(sklearn.neighbors.VALID_METRICS['%s']) to get valid options. Metric can also be a callable function.RNs\Parameter p is found in metric_params. The corresponding parameter from __init__ is ignored.t stackleveliR,R$is/p must be greater than one for minkowski metric(RGRRR(R,R$( RJR1RLR0t VALID_METRICSRMR/twarningstwarnt SyntaxWarningRN(RQt alg_checkt effective_p((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRP|s4        cCsC|jƒ|jdkr%i|_n|jjƒ|_|jjd|jƒ}|jdkrn||jd 0. Got %ds7n_neighbors does not take %s value, enter integer value(R,R$(RGR($RPRMR/teffective_metric_params_tcopytgetRNRLteffective_metric_tpopR1R9tinft isinstanceRFt_fit_Xt_treet _fit_methodRtdataRR tshapeRRJRVRWtVALID_METRICS_SPARSER0RHRURKt issubdtypettypetintegert TypeError(RQtXRZRNt n_samples((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt_fit¤s¨                               cCs |jdkS(NRS(RL(RQ((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt _pairwisesN( t__name__t __module__t__doc__RR/RRRPRrtpropertyRs(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRFjs  ( tcCs|j|||ƒS(s›Helper for the Parallel calls in KNeighborsMixin.kneighbors The Cython method tree.query is not directly picklable by cloudpickle under PyPy. (tquery(ttreeRiRHtreturn_distance((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt_tree_query_parallel_helperstKNeighborsMixincBs;eZdZd„Zdded„Zdddd„ZRS(sMixin for k-neighbors searchescCsÛtj|jdƒdd…df}tj||dddƒ}|dd…d|…f}||tj|||fƒf}|rÑ|jdkr¸tj|||fƒ|f}q×|||f|f}n|}|S(sWReduce a chunk of distances to the nearest neighbors Callback to :func:`sklearn.metrics.pairwise.pairwise_distances_chunked` Parameters ---------- dist : array of shape (n_samples_chunk, n_samples) start : int The index in X which the first row of dist corresponds to. n_neighbors : int return_distance : bool Returns ------- dist : array of shape (n_samples_chunk, n_neighbors), optional Returned only if return_distance neigh : array of shape (n_samples_chunk, n_neighbors) iNiR7R\(R9tarangeRjR/t argpartitiontargsortRbtsqrt(RQR@tstartRHRzt sample_ranget neigh_indtresult((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt_kneighbors_reduce_func*s& "c s§tˆdƒˆd kr%ˆj‰nSˆdkrDtdˆƒ‚n4tjtˆƒtjƒsxtdtˆƒƒ‚nˆd k rŸt }t ˆddƒ‰nt }ˆj ‰ˆd7‰ˆj j d}ˆ|krítd|ˆfƒ‚nˆj \}}tj|ƒd d …d f}tˆjƒ} ˆjd kr¬tˆjd ˆd ˆƒ} ˆjd krpit d6nˆj} ttˆˆj d| dˆjd| | ƒ} nûˆjdkr›tˆƒrÝtdˆjƒ‚nttƒtdƒk} tjd ks | r>| rt nd }ttd|ƒ‰idd6}nttƒ‰idd6}t| |‡‡‡‡‡fd†t ˆj d| ƒDƒƒ} n tdƒ‚ˆràt!| Œ\}}tj"|ƒtj"|ƒf} ntj"| ƒ} |sù| Sˆr| \}}n| }||k}tj#|ddƒ}t |d d …df|>> samples = [[0., 0., 0.], [0., .5, 0.], [1., 1., .5]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(n_neighbors=1) >>> neigh.fit(samples) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> print(neigh.kneighbors([[1., 1., 1.]])) # doctest: +ELLIPSIS (array([[0.5]]), array([[2]])) As you can see, it returns [[0.5]], and [[2]], which means that the element is at distance 0.5 and is the third element of samples (indexes start at 0). You can also query for multiple points: >>> X = [[0., 1., 0.], [1., 0., 1.]] >>> neigh.kneighbors(X, return_distance=False) # doctest: +ELLIPSIS array([[1], [2]]...) Rhis Expected n_neighbors > 0. Got %ds7n_neighbors does not take %s value, enter integer valueR]R^isHExpected n_neighbors <= n_samples, but n_samples = %d, n_neighbors = %dNRRHRzR\tsquaredt reduce_funcRLRORRsQ%s does not work with sparse matrices. Densify the data, or set algorithm='brute's0.12it check_picklet threadingtbackendtthreadstpreferc3s+|]!}ˆˆjˆ|ˆˆƒVqdS(N(Rg(t.0ts(Rpt delayed_queryRHRzRQ(s5lib/python2.7/site-packages/sklearn/neighbors/base.pys Åss$internal: _fit_method not recognizedR7(RR(i(%RR/RHR1R9RlRmRnRotFalseR tTrueRfRjR}RRORhRR…RbR_tlistR RRtjoblib_versiontsyst version_infoRR{RR tziptvstacktalltreshape(RQRpRHRztquery_is_traint train_sizeRqt_R‚ROR‡tkwdsR„t old_joblibRˆtparallel_kwargsR@Rƒt sample_maskt dup_gr_nbrs((RpRRHRzRQs5lib/python2.7/site-packages/sklearn/neighbors/base.pyt kneighborsMsŽ4                     !  t connectivityc Cs:|d kr|j}n|d k rFt|ddƒ}|jd}n|jjd}|jjd}||}tjd|d|ƒ}|dkrÃtj||ƒ}|j||dt ƒ} nL|dkrÿ|j||dt ƒ\}} tj |ƒ}nt d|ƒ‚t || j ƒ|fd ||fƒ} | S( s,Computes the (weighted) graph of k-Neighbors for points in X Parameters ---------- X : array-like, shape (n_query, n_features), or (n_query, n_indexed) if metric == 'precomputed' The query point or points. If not provided, neighbors of each indexed point are returned. In this case, the query point is not considered its own neighbor. n_neighbors : int Number of neighbors for each sample. (default is value passed to the constructor). mode : {'connectivity', 'distance'}, optional Type of returned matrix: 'connectivity' will return the connectivity matrix with ones and zeros, in 'distance' the edges are Euclidean distance between points. Returns ------- A : sparse matrix in CSR format, shape = [n_samples, n_samples_fit] n_samples_fit is the number of samples in the fitted data A[i, j] is assigned the weight of edge that connects i to j. Examples -------- >>> X = [[0], [3], [1]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(n_neighbors=2) >>> neigh.fit(X) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> A = neigh.kneighbors_graph(X) >>> A.toarray() array([[1., 0., 1.], [0., 1., 1.], [1., 0., 1.]]) See also -------- NearestNeighbors.radius_neighbors_graph R]R^iiR£RzR.sRUnsupported mode, must be one of "connectivity" or "distance" but got "%s" insteadRjN(R/RHR RjRfR9R}tonesR¢RR‘travelR1R( RQRpRHtmodet n_samples1t n_samples2t n_nonzerotA_indptrtA_datatA_indtkneighbors_graph((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyR­ïs,,       N(RtRuRvR…R/R‘R¢R­(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyR|'s  #¢cCs|j|||ƒS(s­Helper for the Parallel calls in RadiusNeighborsMixin.radius_neighbors The Cython method tree.query_radius is not directly picklable by cloudpickle under PyPy. (t query_radius(RyRiRIRz((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt"_tree_query_radius_parallel_helper>stRadiusNeighborsMixincBs;eZdZd„Zdded„Zdddd„ZRS(s)Mixin for radius-based neighbors searchesc CsÀg|D]}tj||kƒd^q}|r¶|jdkrzgt|ƒD]#\}}tj|||ƒ^qN}n-gt|ƒD]\}}|||^q‡}||f}n|}|S(s<Reduce a chunk of distances to the nearest neighbors Callback to :func:`sklearn.metrics.pairwise.pairwise_distances_chunked` Parameters ---------- dist : array of shape (n_samples_chunk, n_samples) start : int The index in X which the first row of dist corresponds to. radius : float return_distance : bool Returns ------- dist : list of n_samples_chunk 1d arrays, optional Returned only if return_distance neigh : list of n_samples_chunk 1d arrays iR\(R9twhereRbR;R€( RQR@RRIRztdRƒtitresults((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyt_radius_neighbors_reduce_funcJs,6*c satˆdƒˆdk r4t}tˆddƒ‰nt}ˆj‰ˆdkr[ˆj‰nˆjdkr¡ˆjdkr“ˆˆ9‰itd6}n ˆj }t ˆj dˆdˆƒ}t ˆˆjd |d ˆjd ˆj |}ˆrmt|Œ\}} t|gƒ} t| gƒ} tjt| ƒd d ƒ} | | (tjt| ƒd d ƒ} | | (| | f}qÒt|gƒ} tjt| ƒd d ƒ}| |(n1ˆjdkrÆtˆƒrÒtdˆjƒ‚ntˆj ƒ}ttƒtdƒkrttdtƒ‰idd6}nttƒ‰idd6}t||‡‡‡‡‡fd†tˆjd|ƒDƒƒ}ˆr´tt|Œƒ\} } tj| ƒtj| ƒf}qÒtj|ƒ}n tdƒ‚|sÜ|Sˆrñ|\} } n|} xOt| ƒD]A\}}||k}||| |<ˆr| ||| |>> import numpy as np >>> samples = [[0., 0., 0.], [0., .5, 0.], [1., 1., .5]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(radius=1.6) >>> neigh.fit(samples) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> rng = neigh.radius_neighbors([[1., 1., 1.]]) >>> print(np.asarray(rng[0][0])) # doctest: +ELLIPSIS [1.5 0.5] >>> print(np.asarray(rng[1][0])) # doctest: +ELLIPSIS [1 2] The first array returned contains the distances to all points which are closer than 1.6, while the second array returned contains their indices. In general, multiple points can be queried at the same time. Notes ----- Because the number of neighbors of each point is not necessarily equal, the results for multiple query points cannot be fit in a standard data array. For efficiency, `radius_neighbors` returns arrays of objects, where each object is a 1D array of indices or distances. RhR]R^RR\R†RIRzR‡RLROR8R:RRsQ%s does not work with sparse matrices. Densify the data, or set algorithm='brute's0.12RˆR‰RŠR‹RŒc3s+|]!}ˆˆjˆ|ˆˆƒVqdS(N(Rg(RRŽ(RpRRIRzRQ(s5lib/python2.7/site-packages/sklearn/neighbors/base.pys êsis$internal: _fit_method not recognizedN(RR( RR/RR R‘RfRIRhRbR_RRµR ROR–tsumR9temptytlenRR1RRR“RR¯RR RjttuplethstackR;(RQRpRIRzRšRR‡R´t dist_chunkstneigh_ind_chunkst dist_listtneigh_ind_listR@RƒRORŸtindt ind_neighbortmask((RpRRIRzRQs5lib/python2.7/site-packages/sklearn/neighbors/base.pytradius_neighborsls‚B                 !   R£c Cs‹|d k r*t|ddddgƒ}n|jjd}|d krR|j}n|dkr|j||dtƒ}d }nR|dkrÁ|j||dtƒ\}}tj t |ƒƒ}nt d |ƒ‚|jd}tj g|D]} t | ƒ^qëƒ} tj t |ƒƒ}|d kr?tjt |ƒƒ}ntj tjd d tƒtj| ƒfƒ} t||| fd ||fƒS(såComputes the (weighted) graph of Neighbors for points in X Neighborhoods are restricted the points at a distance lower than radius. Parameters ---------- X : array-like, shape = [n_samples, n_features], optional The query point or points. If not provided, neighbors of each indexed point are returned. In this case, the query point is not considered its own neighbor. radius : float Radius of neighborhoods. (default is the value passed to the constructor). mode : {'connectivity', 'distance'}, optional Type of returned matrix: 'connectivity' will return the connectivity matrix with ones and zeros, in 'distance' the edges are Euclidean distance between points. Returns ------- A : sparse matrix in CSR format, shape = [n_samples, n_samples] A[i, j] is assigned the weight of edge that connects i to j. Examples -------- >>> X = [[0], [3], [1]] >>> from sklearn.neighbors import NearestNeighbors >>> neigh = NearestNeighbors(radius=1.5) >>> neigh.fit(X) # doctest: +ELLIPSIS NearestNeighbors(algorithm='auto', leaf_size=30, ...) >>> A = neigh.radius_neighbors_graph(X) >>> A.toarray() array([[1., 0., 1.], [0., 1., 0.], [1., 0., 1.]]) See also -------- kneighbors_graph R]R^tcsctcooiR£RzR.sQUnsupported mode, must be one of "connectivity", or "distance" but got %s insteadiR8RjN(R/R RfRjRIRÂRR‘R9t concatenateR’R1tarrayR¸R¤tzerostinttcumsumR( RQRpRIR¦R¨R¬R«R@R§taRHRª((s5lib/python2.7/site-packages/sklearn/neighbors/base.pytradius_neighbors_graph s2,         ( N(RtRuRvRµR/R‘RÂRË(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyR°Gs "ŸtSupervisedFloatMixincBseZd„ZRS(cCsLt|ttfƒs6t||ddtƒ\}}n||_|j|ƒS(sÊFit the model using X as training data and y as target values Parameters ---------- X : {array-like, sparse matrix, BallTree, KDTree} Training data. If array or matrix, shape [n_samples, n_features], or [n_samples, n_samples] if metric='precomputed'. y : {array-like, sparse matrix} Target values, array of float values, shape = [n_samples] or [n_samples, n_outputs] R^t multi_output(ReRRR R‘t_yRr(RQRpty((s5lib/python2.7/site-packages/sklearn/neighbors/base.pytfitYs ! (RtRuRÐ(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRÌXstSupervisedIntegerMixincBseZd„ZRS(cCsŽt|ttfƒs6t||ddtƒ\}}n|jdksg|jdkrª|jddkrª|jdkrtjdt ddƒnt |_ |j d ƒ}n t|_ t |ƒg|_tj|jdtjƒ|_xlt|jjdƒD]T}tj|d d …|fd tƒ\}|jd d …|f<|jj|ƒqûW|j s|jd |_|jjƒ|_n|j|ƒS( s©Fit the model using X as training data and y as target values Parameters ---------- X : {array-like, sparse matrix, BallTree, KDTree} Training data. If array or matrix, shape [n_samples, n_features], or [n_samples, n_samples] if metric='precomputed'. y : {array-like, sparse matrix} Target values of shape = [n_samples] or [n_samples, n_outputs] R^RÍiis„A column-vector y was passed when a 1d array was expected. Please change the shape of y to (n_samples, ), for example using ravel().RTiÿÿÿÿR8Ntreturn_inversei(iÿÿÿÿi(ReRRR R‘tndimRjRVRWRRt outputs_2d_R™Rtclasses_R9R·RÈRÎtrangetuniquetappendR¥Rr(RQRpRÏtktclasses((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRÐms& !1     > (RtRuRÐ(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRÑlstUnsupervisedMixincBseZdd„ZRS(cCs |j|ƒS(sFit the model using X as training data Parameters ---------- X : {array-like, sparse matrix, BallTree, KDTree} Training data. If array or matrix, shape [n_samples, n_features], or [n_samples, n_samples] if metric='precomputed'. (Rr(RQRpRÏ((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRИs N(RtRuR/RÐ(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pyRÛ—s(=Rvt functoolsRtdistutils.versionRR”RVtabcRRtnumpyR9t scipy.sparseRRRRRRtbaseRtmetricsR tmetrics.pairwiseR tutilsR R R tutils.multiclassRtutils.validationRt externalsRt exceptionsRt utils._joblibRRRRR“tdictt valid_metricsR’tkeysRURkR3REtwith_metaclassRFR{R:R|R¯R°RÌRÑRÛ(((s5lib/python2.7/site-packages/sklearn/neighbors/base.pytsX           ,"´ ÿ ÿ+