U ֽ|e=1@sddlmZddlZddlZddlZddlmZddl m Z ddl m Z ddlmZmZddlmZmZdifd d Zdifd d Zdifd dZdS))warnN)SpectralEmbedding)pairwise_distances)_VALID_METRICS)pairwise_special_metricSPECIAL_METRICS)SPARSE_SPECIAL_METRICSsparse_named_distances euclideancCs<|dkrtjj||fddStj||jdftjd}|dkrtj||ftjd}|dd} | dkrptj} n,| d krtj} n| d krtj } n t d | t |D]T} ||| k} t | d|D]4} | | dd|| kf} | || | f<| || | f<qqn t |D]}|||kjd d ||<qt j |rDtd|}|tkr^t|||d}n|tkr|t|t||d}nt|rt j |rddtttt@D}z ||}Wntk rtdYnXt|fd|i|}nt|fd|i|}t|d }t|d|d|}||}|S)aProvide a layout relating the separate connected components. This is done by taking the centroid of each component and then performing a spectral embedding of the centroids. Parameters ---------- data: array of shape (n_samples, n_features) The source data -- required so we can generate centroids for each connected component of the graph. n_components: int The number of distinct components to be layed out. component_labels: array of shape (n_samples) For each vertex in the graph the label of the component to which the vertex belongs. dim: int The chosen embedding dimension. metric: string or callable (optional, default 'euclidean') The metric used to measure distances among the source data points. metric_kwds: dict (optional, default {}) Keyword arguments to be passed to the metric function. If metric is 'precomputed', 'linkage' keyword can be used to specify 'average', 'complete', or 'single' linkage. Default is 'average' Returns ------- component_embedding: array of shape (n_components, dim) The ``dim``-dimensional embedding of the ``n_components``-many connected components. Nsize$@dtype precomputedlinkageaveragecompletesinglezPUnrecognized linkage '%s'. Please choose from 'average', 'complete', or 'single'raxiszlForcing component centroids to dense; if you are running out of memory then consider increasing n_neighbors.)metrickwdscSsi|]}t||qS)r ).0krrJ/var/www/website-v5/atlas_env/lib/python3.8/site-packages/umap/spectral.py psz$component_layout..zPMulticomponent layout for custom sparse metrics is not implemented at this time.r) n_componentsZaffinity random_state)nprandomemptyshapefloat64zerosgetmeanmaxmin ValueErrorrangescipysparse isspmatrixrtoarrayrrrcallablesetSKLEARN_PAIRWISE_VALID_METRICSr keysKeyErrorNotImplementedErrorrexpr fit_transform)datar component_labelsdimr!r metric_kwdsZcomponent_centroidsdistance_matrixrc_iZdm_iZc_jdistlabelZfunction_to_name_mapping metric_nameZaffinity_matrixcomponent_embeddingrrrcomponent_layouts+           rDc Cstj|jd|ftjd}|d|kr>t|||||||d} nLtt|d} tt| t | || fg} t | | gd|} t |D]} | || kddf } | dd|| kf} t| | g| }||dkd}| jdd|ks| jd|dkrF|j| || jd|fd | | ||| k<qt| jdd }tjj| jdtjd}tjd t|d| jd| jd}||| |}|d} td| dtt| jd}ztjjj|| d |d t|jd|jddd\}}t|d| }|dd|f}|tt|}||9}|| | ||| k<Wqtjjj k rt!d|j| || jd|fd | | ||| k<YqXq|S)a\Specialised layout algorithm for dealing with graphs with many connected components. This will first fid relative positions for the components by spectrally embedding their centroids, then spectrally embed each individual connected component positioning them according to the centroid embeddings. This provides a decent embedding of each component while placing the components in good relative positions to one another. Parameters ---------- data: array of shape (n_samples, n_features) The source data -- required so we can generate centroids for each connected component of the graph. graph: sparse matrix The adjacency matrix of the graph to be emebdded. n_components: int The number of distinct components to be layed out. component_labels: array of shape (n_samples) For each vertex in the graph the label of the component to which the vertex belongs. dim: int The chosen embedding dimension. metric: string or callable (optional, default 'euclidean') The metric used to measure distances among the source data points. metric_kwds: dict (optional, default {}) Keyword arguments to be passed to the metric function. Returns ------- embedding: array of shape (n_samples, dim) The initial embedding of ``graph``. rrrrr=g@Ngrlowhighr r?SM-C6?whichncvtolv0maxiterWARNING: spectral initialisation failed! The eigenvector solver failed. This is likely due to too small an eigengap. Consider adding some noise or jitter to your data. Falling back to random initialisation!)"r"r$r%float32rDintceilhstackeyer'vstackr-tocsrtocsctocoorr+uniformasarraysumr.r/identityr&spdiagssqrtr*linalgeigshonesargsortabs ArpackErrorr)r:graphr r;r<r!rr=resultZmeta_embeddingrbaserAZcomponent_graph distances data_range diag_dataIDLnum_lanczos_vectors eigenvalues eigenvectorsorderrC expansionrrrmulti_component_layouts0  "(   "    rwc Cs|jd}tjj|\}}|dkr)largestrPNrSg$r rF)r%r.r/csgraphconnected_componentsrwr"r^r_r`r&rarbr*rUrcrdrelobpcgnormalrfrhrr])r:rir<r!rr= n_samplesr labelsrnrorprqrrrrsrtrurrrspectral_layouts`   "  r)warningsrnumpyr" scipy.sparser.scipy.sparse.csgraphZsklearn.manifoldrsklearn.metricsrZsklearn.metrics.pairwiserr4umap.distancesrr umap.sparserr rDrwrrrrrs"