a Kb7'@sddlZddlmZddlmZddlmZz ddlZWneyJYn0ddl m Z Gddde Z dd Z d d Z Gd d d ZGdddZdS)N)abstractmethod)sqrt)stdout)ClusterIc@sdeZdZdZdddZdddZedd Zd d Zed d Z ddZ ddZ ddZ ddZ dS)VectorSpaceClustererz Abstract clusterer which takes tokens and maps them into a vector space. Optionally performs singular value decomposition to reduce the dimensionality. FNcCsd|_||_||_dS)a) :param normalise: should vectors be normalised to length 1 :type normalise: boolean :param svd_dimensions: number of dimensions to use in reducing vector dimensionsionality with SVD :type svd_dimensions: int N)_Tt_should_normalise_svd_dimensions)selfZ normaliseZsvd_dimensionsr 0lib/python3.9/site-packages/nltk/cluster/util.py__init__szVectorSpaceClusterer.__init__c st|dksJjr&ttj|}jrĈjt|dkrtjt t |\}}}|djt jtj }|dddjf}|djddf} t t || }t |_|||rfdd|DSdS)Nrcsg|]}|qSr )classify).0vectorr r r ?z0VectorSpaceClusterer.cluster..)lenrlistmap _normaliser numpyZlinalgZsvdZ transposeZarrayidentityZfloat64dotrcluster_vectorspace) r vectorsZassign_clusterstraceudZvtSTZDtr rr cluster(s  zVectorSpaceClusterer.clustercCsdS)zD Finds the clusters using the given set of vectors. Nr )r rrr r r rAsz(VectorSpaceClusterer.cluster_vectorspacecCs<|jr||}|jdur(t|j|}||}||SN)rrrrrclassify_vectorspaceZ cluster_name)r rr"r r r rGs    zVectorSpaceClusterer.classifycCsdS)zN Returns the index of the appropriate cluster for the vector. Nr r rr r r r$Osz)VectorSpaceClusterer.classify_vectorspacecCs4|jr||}|jdur(t|j|}|||Sr#)rrrrrlikelihood_vectorspace)r rZlabelr r r likelihoodUs   zVectorSpaceClusterer.likelihoodcCs||}||krdSdS)zP Returns the likelihood of the vector belonging to the cluster. g?g)r$)r rr"Z predictedr r r r&\s z+VectorSpaceClusterer.likelihood_vectorspacecCs,|jr||}|jdur(t|j|}|S)zU Returns the vector after normalisation and dimensionality reduction N)rrrrrr%r r r rcs   zVectorSpaceClusterer.vectorcCs|tt||S)z7 Normalises the vector to unit length. rrrr%r r r rmszVectorSpaceClusterer._normalise)FN)FF)__name__ __module__ __qualname____doc__r r"rrrr$r'r&rrr r r r rs    rcCs||}tt||S)z} Returns the euclidean distance between vectors u and v. This is equivalent to the length of the vector (u - v). r()rvdiffr r r euclidean_distancetsr/cCs0dt||tt||tt||S)zw Returns 1 minus the cosine of the angle between vectors v and u. This is equal to ``1 - (u.v / |u||v|)``. )rrr)rr-r r r cosine_distance}sr1c@s2eZdZdZddZd ddZddZd d Zd S) _DendrogramNodezTree node of a dendrogram.cGs||_||_dSr#)_value _children)r valueZchildrenr r r r sz_DendrogramNode.__init__TcCs@|jr*g}|jD]}|||q|S|r6|jgS|gSdSr#)r4extendleavesr3)r valuesr7childr r r r7s z_DendrogramNode.leavescCs|j|fg}t||krv|\}}|js:|||fqv|jD]*}|jr\||j|fq@|d|fq@|q g}|D]\}}||q~|SNr)r3rpopr4pushappendsortr7)r nqueueprioritynoder9groupsr r r rCs      z_DendrogramNode.groupscCst|j|jdkSr:)r1r3)r Z comparatorr r r __lt__sz_DendrogramNode.__lt__N)T)r)r*r+r,r r7rCrDr r r r r2s  r2c@s@eZdZdZgfddZddZddZgfdd Zd d Zd S) Dendrograma Represents a dendrogram, a tree with a specified branching order. This must be initialised with the leaf items, then iteratively call merge for each branch. This class constructs a tree representing the order of calls to the merge function. cCs(dd|D|_t|j|_d|_dS)zs :param items: the items at the leaves of the dendrogram :type items: sequence of (any) cSsg|] }t|qSr )r2ritemr r r rrz'Dendrogram.__init__..r0N)_itemscopy_original_items_merge)r itemsr r r r szDendrogram.__init__csjt|dksJtjgfdd|DR}jd7_|j|d<|ddD] }j|=qXdS)a= Merges nodes at given indices in the dendrogram. The nodes will be combined which then replaces the first node specified. All other nodes involved in the merge will be removed. :param indices: indices of the items to merge (at least two) :type indices: seq of int c3s|]}j|VqdSr#)rH)rirr r rz#Dendrogram.merge..r0rN)rr2rKrH)r indicesrBrNr rr merges  zDendrogram.mergecCs8t|jdkr$t|jg|jR}n |jd}||S)z Finds the n-groups of items (leaves) reachable from a cut at depth n. :param n: number of groups :type n: int r0r)rrHr2rKrC)r r?rootr r r rCs zDendrogram.groupscs0d\}}}t|jdkr.t|jg|jR}n |jd}|j}|rH|}ndd|D}ttt|ddtddfdd d d }|j|fg} fd d|D} | r| \} } t td d| j } t t|j | }| rt |}t|}tt|D]}||| vrv||kr<||d|n,||krX|||dn|||||| |<n:||kr|krnn||||n || |q|d| j D]}|j r| |j|fq| | D]}||q|dq|dfdd|D|ddS)z Print the dendrogram in ASCII art to standard out. :param leaf_labels: an optional list of strings to use for labeling the leaves :type leaf_labels: list )+-|r0rcSsg|]}d|jqS)z%s)r3rZleafr r r rrz#Dendrogram.show..rM cs|||Sr#r )Zcentreleftright)lhalfrhalfr r formatszDendrogram.show..formatcSst|dSr#)rwrite)strr r r displaysz Dendrogram.show..displaycsg|] }dqS)rWr rV)r\r r rrcSs|ddS)NFr)r7)cr r r rz!Dendrogram.show.. c3s|]}|VqdSr#)centerrF)widthr r rO#rz"Dendrogram.show..N)rWrW)rrHr2rKrJmaxrintr3r;rr4indexminranger=r>join)r Z leaf_labelsZJOINZHLINKZVLINKrRr7Zlast_rowr_r@Z verticalsrArBZchild_left_leafrPZmin_idxZmax_idxrNr9Zverticalr )r\rZr[rer showsV         zDendrogram.showcCsDt|jdkr$t|jg|jR}n |jd}|d}dt|S)Nr0rFz)rrHr2rKr7)r rRr7r r r __repr__&s   zDendrogram.__repr__N) r)r*r+r,r rQrCrlrmr r r r rEs   IrE)rIabcrZmathrsysrr ImportErrorZnltk.cluster.apirrr/r1r2rEr r r r s      _ +