ó ‡ˆ\c@sldZddlZddlmZddlmZddlmZddl m Z defd „ƒYZ dS( sY Feature agglomeration. Base classes and functions for performing feature agglomeration. iÿÿÿÿNi(tTransformerMixin(t check_array(tcheck_is_fitted(tissparsetAgglomerationTransformcBs)eZdZejZd„Zd„ZRS(sG A class for feature agglomeration via the transform interface c CsAt|dƒ|j}t|ƒ}t|jƒ|jdkrMtdƒ‚n|tjkrÖt |ƒ rÖtj |jƒ}|jd}tj gt |ƒD]/}tj |j||dd…fƒ|^q›ƒ}ngg}xLtj |jƒD]8}|j||dd…|j|kfddƒƒqïWtj |ƒj}|S(sÎ Transform a new matrix using the built clustering Parameters ---------- X : array-like, shape = [n_samples, n_features] or [n_features] A M by N array of M observations in N dimensions or a length M array of M one-dimensional observations. Returns ------- Y : array, shape = [n_samples, n_clusters] or [n_clusters] The pooled values for each feature cluster. tlabels_is9X has a different number of features than during fitting.iNtaxis(Rt pooling_funcRtlenRtshapet ValueErrortnptmeanRtbincounttarraytrangetuniquetappendtT(tselftXRtsizet n_samplestitnXtl((sElib/python2.7/site-packages/sklearn/cluster/_feature_agglomeration.pyt transforms     E6cCs9t|dƒtj|jdtƒ\}}|d|fS(s' Inverse the transformation. Return a vector of size nb_features with the values of Xred assigned to each group of features Parameters ---------- Xred : array-like, shape=[n_samples, n_clusters] or [n_clusters,] The values to be assigned to each cluster of samples Returns ------- X : array, shape=[n_samples, n_features] or [n_features] A vector of size n_samples with the values of Xred assigned to each of the cluster of samples. Rtreturn_inverse.(RR RRtTrue(RtXredtuniltinverse((sElib/python2.7/site-packages/sklearn/cluster/_feature_agglomeration.pytinverse_transform=s (t__name__t __module__t__doc__R R RRR (((sElib/python2.7/site-packages/sklearn/cluster/_feature_agglomeration.pyRs  #( R#tnumpyR tbaseRtutilsRtutils.validationRt scipy.sparseRR(((sElib/python2.7/site-packages/sklearn/cluster/_feature_agglomeration.pyts