+ h &^RIt^RIHt^RIHt^RIHt^RIt^RI H t !RR] 4t Rt R t !R R 4t!R R 4tR# ]dL6i;i)N)abstractmethod)sqrt)stdout)ClusterIcva]tRt^toRtR RltRRlt]R4tRt ]R4t Rt R t R t R tR tVtR#)VectorSpaceClustererz Abstract clusterer which takes tokens and maps them into a vector space. Optionally performs singular value decomposition to reduce the dimensionality. Nc ,RVnWnW nR#)z :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)self normalisesvd_dimensionss&&&3/usr/lib/python3/dist-packages/nltk/cluster/util.py__init__VectorSpaceClusterer.__init__s!*-cr\V4^8gQhVP'd \\VPV44pVP 'Ed#VP \V^,48Ed\ PP\ P!\ P!V444wrEpVRVP \ P!VP \ P4,pVRRVP 13,pVRVP 1R3,p \ P!\ P!Wy44p\ P!V4VnVPW4V'd!VU u.uFqP!V 4NK up #R#uup i)rN:NNN)lenr listmap _normaliser numpylinalgsvd transposearrayidentityfloat64dotr cluster_vectorspaceclassify) rvectorsassign_clusterstraceudvtSTDtvectors &&&& rclusterVectorSpaceClusterer.cluster(sN7|a  ! ! !3t89G    D$8$83wqz?$J))%//%++g:N*OPJQ2(D(()ENN$$emm-A!+t++++,A*d***A-.Booeii&67Gq)DH   0 8?@fMM&)@ @ @sF4c R#)z4 Finds the clusters using the given set of vectors. N)rr$r&s&&&rr"(VectorSpaceClusterer.cluster_vectorspaceArcVP'dVPV4pVPe"\P!VPV4pVP V4pVP V4#N)r rr rr!classify_vectorspace cluster_name)rr-r.s&& rr#VectorSpaceClusterer.classifyGs[  ! ! !__V,F 88 YYtxx0F++F3  ))rc R#)z> Returns the index of the appropriate cluster for the vector. Nr1rr-s&&rr6)VectorSpaceClusterer.classify_vectorspaceOr3rcVP'dVPV4pVPe"\P!VPV4pVP W4#r5)r rr rr!likelihood_vectorspace)rr-labels&&&r likelihoodVectorSpaceClusterer.likelihoodUsK  ! ! !__V,F 88 YYtxx0F**699rc 8VPV4pW#8XdR#R#)z@ Returns the likelihood of the vector belonging to the cluster. g?g)r6)rr-r. predicteds&&& rr=+VectorSpaceClusterer.likelihood_vectorspace\s$--f5 *s33rc VP'dVPV4pVPe"\P!VPV4pV#)zE Returns the vector after normalisation and dimensionality reduction )r rr rr!r:s&&rr-VectorSpaceClusterer.vectorcsA  ! ! !__V,F 88 YYtxx0F rc NV\\P!W44, #)z' Normalises the vector to unit length. rrr!r:s&&rrVectorSpaceClusterer._normalisemsUYYv6777rr )FN)FF)__name__ __module__ __qualname____firstlineno____doc__rr.rr"r#r6r?r=r-r__static_attributes____classdictcell__ __classdict__s@rrrs[ .A2  *  :488rrcPW, p\\P!W"44#)zq Returns the euclidean distance between vectors u and v. This is equivalent to the length of the vector (u - v). rG)r'vdiffs&& reuclidean_distancerUts 5D  $% &&rc ^\P!W4\\P!W44\\P!W44,, , #)zk Returns 1 minus the cosine of the angle between vectors v and u. This is equal to ``1 - (u.v / |u||v|)``. )rr!r)r'rSs&&rcosine_distancerW}s;  !4 !#84 !;P#PQ RRrc@a]tRt^toRtRtRRltRtRtRt Vt R#) _DendrogramNodezTree node of a dendrogram.cWnW nR#r5)_value _children)rvaluechildrens&&*rr_DendrogramNode.__init__s  !rcVP'd8.pVPF#pVPVPV44K% V#V'dVP.#V.#r5)r\extendleavesr[)rvaluesrbchilds&& rrb_DendrogramNode.leavessN >>>F ell623(M KK= 6MrcVPV3.p\V4V8dVP4wr4VP'gVP W434MiVPFGpVP'd VP VPV34K4VP ^V34KI VP 4K.pVF$wr4VP VP44K& V#r)r[rpopr\pushappendsortrb)rnqueueprioritynoderdgroupss&& rrp_DendrogramNode.groupss++t$%%j1n"YY[NH>>> H+,???LL%,,!67LL!U, ( JJL#NH MM$++- ($ rcH\VPVP4^8#rg)rWr[)r comparators&&r__lt___DendrogramNode.__lt__st{{J,=,=>BBr)r\r[N)T) rIrJrKrLrMrrbrprtrNrOrPs@rrYrYs#$" *CCrrYcNa]tRt^toRt.3RltRtRt.3RltRt Rt Vt R#) Dendrogramz 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. c VUu.uFp\V4NK upVn\P!VP4Vn^VnR#uupi)z[ :param items: the items at the leaves of the dendrogram :type items: sequence of (any) N)rY_itemscopy_original_items_merge)ritemsitems&& rrDendrogram.__init__sA :??t,? #yy5 @sAc a\V4^8gQh\SP.V3RlV4O5!pS;P^, unVSPV^,&VR,FpSPVK R#)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 c3J<"TFpSPV,xK R#5ir5)ry).0irs& r #Dendrogram.merge..s-Ngdkk!nngs #:NNN)rrYr|ry)rindicesrorsf* rmergeDendrogram.mergesk7|q   t{{O-Ng-NO q "& GAJA Arc \VP4^8d#\VP.VPO5!pMVP^,pVP V4#)zr Finds the n-groups of items (leaves) reachable from a cut at depth n. :param n: number of groups :type n: int )rryrYr|rp)rrlroots&& rrpDendrogram.groupssG t{{ a "4;;==D;;q>D{{1~rc aaaRRRrCp\VP4^8d#\VP.VPO5!pMVP^,pVPpV'dTpM"VUu.uFpRVP ,NK pp\ \\V44^,oS^,o\SS, ^, 4oR VV3Rllp Rp VP V3.p VUu.uF q!R4NK p pV 'EdV P4wr\\RVP44p\\VPV44pV'd\V4p\ V4p\\V44FpVV,V9dLVX8XdV !V !VRV44M&VX8XdV !V !W#R44MV !V !W#V44V !V4V V&K\XTu;8:dX8:dMMV !V !W3V44KV !V V,4K V !R 4VPF4pVP'gKV PVP V34K6 V P!4V F pV !V4K V !R 4EKV !R P#V3R lV444V !R 4R #uupiuupi)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 +-|z%s c0<SV, V VS, 2#r5r1)centreleftrightlhalfrhalfs&&&rformatDendrogram.show..formatsdl^F8EEM?; ;rc2\P!V4R#r5)rwrite)strs&rdisplay Dendrogram.show..displays LL rc2VPR4^,#)F)rb)cs&r!Dendrogram.show..s%1Cr c3D<"TFqPS4xK R#5ir5)center)rr~widths& rr"Dendrogram.show..#s@xt E**xs N)rr)rryrYr|r{r[maxrintrhrr\indexminrangerjrkjoin)r leaf_labelsJOINHLINKVLINKrrblast_rowleafrrrm verticalsrnrochild_left_leafrmin_idxmax_idxrrdverticalrrrs&& @@@rshowDendrogram.showsO!#sU t{{ a "4;;==D;;q>D%% "H7=>vtt{{**vH>CX&'!+ EEMA%& < < ++t$%-34VTVC[V 4e"YY[NH"3'CT^^#TUO3v||_=>Gg,g,3v;'!9/G|tS% 89gtC 89tE :;#)%=IaL,W,F578IaL)( DM???LL%,,!67( JJL%!& DM @x@@A c? 5s 3K2K c\VP4^8d#\VP.VPO5!pMVP^,pVP R4pR\V4,#)rFz)rryrYr|rb)rrrbs& r__repr__Dendrogram.__repr__&sT t{{ a "4;;==D;;q>DU#,s6{::r)ryr|r{N) rIrJrKrLrMrrrprrrNrOrPs@rrwrws4   "GR;;rrw)rzabcrmathrsysrr ImportErrornltk.cluster.apirrrUrWrYrwr1rrrsh  &\88\8~'S(C(CV|;|;C  sA AA