U 9hO@sddlZddlmZddlmZddlmZddlm Z ddl m Z ddl m Z mZmZdZdZd Zd Zd d Zd dZddZGdddeZddZddZGdddeZGdddeZGdddZdS)N) dendrogram)TSNE)PCA)pairwise_distances)warn)compute_stabilitylabelling_at_cutrecurse_leaf_dfscCs:g}|g}|r6|||dt|d|}q |S)zK Perform a breadth first search on a tree in condensed tree format childparent)extendnpisintolist)treeZbfs_rootresultZ to_processr`/var/www/html/CrowdFlow/HYROX/ble_analysis_env_py38/lib/python3.8/site-packages/hdbscan/plots.py_bfs_from_cluster_trees  rcsBd|kd}t|dkr&|gStfdd|DgSdS)Nrr rcsg|]}t|qSr)r .0r  cluster_treerr (sz%_recurse_leaf_dfs..)lensum)rZ current_nodechildrenrrr_recurse_leaf_dfs#s r cCsB||ddk}|jddkr,|dgS|d}t||S)N child_sizerrr)shapeminr )Zcondensed_treerrootrrr _get_leaves*s  r%c @sNeZdZdZdddZddd Zd d ZdddZddZddZ ddZ d S) CondensedTreeaaThe condensed tree structure, which provides a simplified or smoothed version of the :class:`~hdbscan.plots.SingleLinkageTree`. Parameters ---------- condensed_tree_array : numpy recarray from :class:`~hdbscan.HDBSCAN` The raw numpy rec array version of the condensed tree as produced internally by hdbscan. cluster_selection_method : string, optional (default 'eom') The method of selecting clusters. One of 'eom' or 'leaf' allow_single_cluster : Boolean, optional (default False) Whether to allow the root cluster as the only selected cluster eomFcCs||_||_||_dSN) _raw_treecluster_selection_methodallow_single_cluster)selfZcondensed_tree_arrayr*r+rrr__init__DszCondensedTree.__init__rc'st|j}|jd}|jd}t|tjr<|i}n$tt|fddt t |D}|di}t ||ddD]} |jddg} | |jd| k|jd dk@} t | ddkrx| d\} } t || || g|| <| dd || <| dd|| <qxg} g}g}g}i}t |j|jd|kd }|rDt |}t ||ddD]}d d d d g||<|j|jd|k}t |d }||}|}|d}t ||d|kd }|rt |}t |}t |}t||}||}||||d ||t<||||d ||t<||||t<t|d||t<|}|}t|dD]}||}|d|kr|||ks|d|kr| |||||d||||||}|}|r"t||d } | d krt t||d }nd}n ||d 8}|d}qjqTg}!g}"|j|jd dkD]}|d}#|d}$|d }%|rt |%}%t||$||#}&|!||#|||$||&|%d g|"||$||$gqZ| ||||!|"|d S)aGenerates data for use in plotting the 'icicle plot' or dendrogram plot of the condensed tree generated by HDBSCAN. Parameters ---------- leaf_separation : float, optional How far apart to space the final leaves of the dendrogram. (default 1) log_size : boolean, optional Use log scale for the 'size' of clusters (i.e. number of points in the cluster at a given lambda value). (default False) max_rectangles_per_icicle : int, optional To simplify the plot this method will only emit ``max_rectangles_per_icicle`` bars per branch of the dendrogram. This ensures that we don't suffer from massive overplotting in cases with a lot of data points. Returns ------- plot_data : dict Data associated to bars in a bar plot: `bar_centers` x coordinate centers for bars `bar_tops` heights of bars in lambda scale `bar_bottoms` y coordinate of bottoms of bars `bar_widths` widths of the bars (in x coord scale) `bar_bounds` a 4-tuple of [left, right, bottom, top] giving the bounds on a full set of cluster bars Data associates with cluster splits: `line_xs` x coordinates for horizontal dendrogram lines `line_ys` y coordinates for horizontal dendrogram lines rcsg|] }|qSrrrxleaf_separationrrr{sz/CondensedTree.get_plot_data..grr lambda_valr!r@g{Gz?) bar_centersbar_tops bar_bottoms bar_widthsline_xsline_yscluster_bounds)r%r)maxr# isinstancerZint64dictziprangermeanrlogfloatCB_LEFTCB_RIGHT CB_BOTTOMCB_TOPZargsortappendexpsign)'r,r2log_sizemax_rectangle_per_icicleleavesZ last_leafr$Zcluster_x_coordsZcluster_y_coordsclustersplit left_child right_childr6r7r8r9r<ZscalingcZ c_childrenZ current_sizeZcurrent_lambdaZcluster_max_sizeZcluster_max_lambdaZcluster_min_sizeZtotal_size_changeZstep_size_changeZlast_step_sizeZlast_step_lambdairowZexp_sizer:r;rr r!rKrr1r get_plot_dataJs'                       zCondensedTree.get_plot_datacs|jdkrt|j|jr,tdd}ntdddd}|j|jddk}dd|D|D]n}|d |k}tfd d |d |D}||krd |<||<qjt||D]}||krd |<qqjtfdd DS|jdkrt |jSt ddS)Nr'T)reverser3r!rcSsi|] }|dqS)TrrrOrrr sz2CondensedTree._select_clusters..rcsg|] }|qSrrr) stabilityrrrsz2CondensedTree._select_clusters..r Fcsg|]}|r|qSrrrX) is_clusterrrrsleafzEInvalid Cluster Selection Method: %s Should be one of: "eom", "leaf" ) r*rr)r+sortedkeysrrrr% ValueError)r,Z node_listrnodeZchild_selectionZsubtree_stabilityZsub_noder)r[rZr_select_clusterss0          zCondensedTree._select_clustersviridisNTc s|zddlm} Wntk r,tdYnX|j||| d|dkr| jj|| dtdddfdd dD} nd } |dkr| }|j d d d d| dddg} t ddD]\} }| | | |q|j | d dd|r zddlm}Wntk rBtdYnX|}tfdd |D}t|s|tdtd d g}|t|}tt|t|g}ttj|ddgd}t|D].\}}d|}|t|t}|t|t}t|t|tgt|t|tgf}t|dsX|d|f}t|sh|}d|}||kr~|}|dk rt |t |kr||}nd}||d|d|d|d d!}|r|j!t"|||dd"||dd#|fd$d%d&|#|q|r@| j$|d'}|r4|j%&d(n |j%&d)|'gd*D]}|j(|)d+qN|*|&d,|S)-a Use matplotlib to plot an 'icicle plot' dendrogram of the condensed tree. Effectively this is a dendrogram where the width of each cluster bar is equal to the number of points (or log of the number of points) in the cluster at the given lambda value. Thus bars narrow as points progressively drop out of clusters. The make the effect more apparent the bars are also colored according the the number of points (or log of the number of points). Parameters ---------- leaf_separation : float, optional (default 1) How far apart to space the final leaves of the dendrogram. cmap : string or matplotlib colormap, optional (default viridis) The matplotlib colormap to use to color the cluster bars. select_clusters : boolean, optional (default False) Whether to draw ovals highlighting which cluster bar represent the clusters that were selected by HDBSCAN as the final clusters. label_clusters : boolean, optional (default False) If select_clusters is True then this determines whether to draw text labels on the clusters. selection_palette : list of colors, optional (default None) If not None, and at least as long as the number of clusters, draw ovals in colors iterating through this palette. This can aid in cluster identification when plotting. axis : matplotlib axis or None, optional (default None) The matplotlib axis to render to. If None then a new axis will be generated. The rendered axis will be returned. colorbar : boolean, optional (default True) Whether to draw a matplotlib colorbar displaying the range of cluster sizes as per the colormap. log_size : boolean, optional (default False) Use log scale for the 'size' of clusters (i.e. number of points in the cluster at a given lambda value). max_rectangles_per_icicle : int, optional (default 20) To simplify the plot this method will only emit ``max_rectangles_per_icicle`` bars per branch of the dendrogram. This ensures that we don't suffer from massive overplotting in cases with a lot of data points. Returns ------- axis : matplotlib axis The axis on which the 'icicle plot' has been rendered. rNzYou must install the matplotlib library to plot the condensed tree.Use get_plot_data to calculate the relevant data without plotting.)r2rLrMnoner9cmapZnormcsg|]}|qSr)to_rgbar/)smrrrYsz&CondensedTree.plot..blackr6r7r8center)bottomwidthcoloralign linewidthr:r;r)rlrn)EllipsezEYou must have matplotlib.patches available to plot selected clusters.csg|]}d|qS)r<r)rrS) plot_datarrr}szcInfinite lambda values encountered in chosen clusters. This might be due to duplicates in the data. Z)qr<g?rr5g333333?r )Z facecolor edgecolorrng@g?leftrj)ZxyZxytextZhorizontalalignmentZverticalalignmentaxlog(Number of points)zNumber of pointsrighttoprjFz$\lambda$ value)+matplotlib.pyplotpyplot ImportErrorrVcmScalarMappable Normalizer= set_arraygcabarr@rIplotZmatplotlib.patchesrorararrayisfiniteallrZhstackrBr#diffZ percentile enumeraterFrErHrGrZannotatestr add_artistcolorbarrx set_ylabel set_xticksspines set_visibleZ invert_yaxis) r,r2reZselect_clustersZlabel_clustersZselection_paletteaxisrrLZmax_rectangles_per_iciclepltZ bar_colorsZ drawlinesZxsZysroZchosen_clustersr<Z plot_rangeZ mean_y_center max_heightrTrSZc_boundsrkheightriZ min_heightZ oval_colorboxcbsider)rprgrr s?               zCondensedTree.plotcCs |jS)zNReturn a numpy structured array representation of the condensed tree. )r)copyr,rrrto_numpyszCondensedTree.to_numpycCs@zddlm}m}Wntk r0tdYnX||j}|S)aReturn a pandas dataframe representation of the condensed tree. Each row of the dataframe corresponds to an edge in the tree. The columns of the dataframe are `parent`, `child`, `lambda_val` and `child_size`. The `parent` and `child` are the ids of the parent and child nodes in the tree. Node ids less than the number of points in the original dataset represent individual points, while ids greater than the number of points are clusters. The `lambda_val` value is the value (1/distance) at which the `child` node leaves the cluster. The `child_size` is the number of points in the `child` node. r DataFrameSeries:You must have pandas installed to export pandas DataFrames)pandasrrrr))r,rrrrrr to_pandass  zCondensedTree.to_pandascCs~zddlm}m}Wntk r0tdYnX|}|jD] }|j|d|d|ddq>||t|jddgd |S) aReturn a NetworkX DiGraph object representing the condensed tree. Edge weights in the graph are the lamba values at which child nodes 'leave' the parent cluster. Nodes have a `size` attribute attached giving the number of points that are in the cluster (or 1 if it is a singleton point) at the point of cluster creation (fewer points may be in the cluster at larger lambda values). rDiGraphset_node_attributes:You must have networkx installed to export networkx graphsrr r4weightr!size)networkxrrrr)add_edger?)r,rrrrUrrr to_networkxs  zCondensedTree.to_networkx)r'F)rFr.) rrbFFNNTFr.) __name__ __module__ __qualname____doc__r-rVrarrrrrrrrr&3s,  !! :r&cCsJ||kr gStt|||d||tt|||d|||gS)Nrr)_get_dendrogram_orderingint)rlinkager$rrrrsrcCsg}|D]}|||d|krB|t|||d|d}nd}|||d|kr||t|||d|d}nd}|||fq|S)Nrr r)rrI)Zorderingrr$ linewidthsr0Z left_widthZ right_widthrrr_calculate_linewidthss""rc@sDeZdZdZddZddd Zd d Zd d ZddZdddZ dS)SingleLinkageTreea&A single linkage format dendrogram tree, with plotting functionality and networkX support. Parameters ---------- linkage : ndarray (n_samples, 4) The numpy array that holds the tree structure. As output by scipy.cluster.hierarchy, hdbscan, of fastcluster. cCs ||_dSr()_linkage)r,rrrrr-#szSingleLinkageTree.__init__NrTrbc Csxt|j||dd}|d}|d} zddlm} Wntk rNtdYnX|dkr`| }|rtdt|j|jt|jd } t| |jt|jd } nd gt| } |d krt t |  } | j j|| d| d }|| t|| | D]$\}}}|dd}|dd}|dd}|dd}|d d }|d d }|d kr|j|||t |dt d |dddd|j|||t |d t d |d dddnH|j||dt d |dddd|j||dt d |d ddd|j||dddddq|rD| j||d}|jd|gdD]}|j|dqR|d|S)aPlot a dendrogram of the single linkage tree. Parameters ---------- truncate_mode : str, optional The dendrogram can be hard to read when the original observation matrix from which the linkage is derived is large. Truncation is used to condense the dendrogram. There are several modes: ``None/'none'`` No truncation is performed (Default). ``'lastp'`` The last p non-singleton formed in the linkage are the only non-leaf nodes in the linkage; they correspond to rows Z[n-p-2:end] in Z. All other non-singleton clusters are contracted into leaf nodes. ``'level'/'mtica'`` No more than p levels of the dendrogram tree are displayed. This corresponds to Mathematica(TM) behavior. p : int, optional The ``p`` parameter for ``truncate_mode``. vary_line_width : boolean, optional Draw downward branches of the dendrogram with line thickness that varies depending on the size of the cluster. cmap : string or matplotlib colormap, optional The matplotlib colormap to use to color the cluster bars. A value of 'none' will result in black bars. (default 'viridis') colorbar : boolean, optional Whether to draw a matplotlib colorbar displaying the range of cluster sizes as per the colormap. (default True) Returns ------- axis : matplotlib axis The axis on which the dendrogram plot has been rendered. T)p truncate_modeZno_plotZicoordZdcoordrNzHYou must install the matplotlib library to plot the single linkage tree.r r)?rrcrdr ZmiterZbutt)rlrnZsolid_joinstyleZsolid_capstylekrrwryrzFdistance)rrr}r~rrrrrrlog2rflattenrrrr=rr@rrfrrxrrrr)r,rrrvary_line_widthrerZdendrogram_dataXYrZdendrogram_orderingrZ color_arrayrgr0yZlwZleft_xZright_xZleft_yZright_yZ horizontal_xZ horizontal_yrrrrrr&sv/"              zSingleLinkageTree.plotcCs |jS)a>Return a numpy array representation of the single linkage tree. This representation conforms to the scipy.cluster.hierarchy notion of a single linkage tree, and can be used with all the associated scipy tools. Please see the scipy documentation for more details on the format. )rrrrrrrszSingleLinkageTree.to_numpycCszddlm}m}Wntk r0tdYnXd|jjd}||jjdd}t||d}|||jjd|jjd|jjd|jjdddd d d d g}|S) aReturn a pandas dataframe representation of the single linkage tree. Each row of the dataframe corresponds to an edge in the tree. The columns of the dataframe are `parent`, `left_child`, `right_child`, `distance` and `size`. The `parent`, `left_child` and `right_child` are the ids of the parent and child nodes in the tree. Node ids less than the number of points in the original dataset represent individual points, while ids greater than the number of points are clusters. The `distance` value is the at which the child nodes merge to form the parent node. The `size` is the number of points in the `parent` node. rrrr rr )rrQrRrrrrQrRrr) rrrrrr"rZarangeT)r,rrmax_node num_pointsZ parent_arrayrrrrrs"     zSingleLinkageTree.to_pandasc Cszddlm}m}Wntk r0tdYnXd|jjd}||jjdd}|}t|j|D]8\}}|j||d|dd|j||d|ddqhddt|j|D}|||d |S) a6Return a NetworkX DiGraph object representing the single linkage tree. Edge weights in the graph are the distance values at which child nodes merge to form the parent cluster. Nodes have a `size` attribute attached giving the number of points that are in the cluster. rrrr rrcSsi|]\}}||dqS)r r)rrrUrrrrYsz1SingleLinkageTree.to_networkx..r)rrrrrr"rr) r,rrrrrrrUZ size_dictrrrrs  zSingleLinkageTree.to_networkxcCst|j||S)a?Return a flat clustering from the single linkage hierarchy. This represents the result of selecting a cut value for robust single linkage clustering. The `min_cluster_size` allows the flat clustering to declare noise points (and cluster smaller than `min_cluster_size`). Parameters ---------- cut_distance : float The mutual reachability distance cut value to use to generate a flat clustering. min_cluster_size : int, optional Clusters smaller than this value with be called 'noise' and remain unclustered in the resulting flat clustering. Returns ------- labels : array [n_samples] An array of cluster labels, one per datapoint. Unclustered points are assigned the label -1. )r r)r,Z cut_distanceZmin_cluster_sizerrr get_clustersszSingleLinkageTree.get_clusters)NNrTrbT)r) rrrrr-rrrrrrrrrrs  l %rc @s6eZdZddZdd d Zd dZddZddZdS)MinimumSpanningTreecCs||_||_dSr()_mst_data)r,Zmstdatarrrr-szMinimumSpanningTree.__init__N(r皙?? viridis_rr Tc Cszddlm} ddlm} Wntk r8tdYnX|jjddkrVtddS|dkrf| }|jjddkr|jjdd krt d d  |j} n |j } t | } n |j } |r|t |jjd|jjdd }n|}| |jddddft}| ||||d }||jdddfj|||j| jd| jd|||d |g|g| r| j||d}|jd|S)aPlot the minimum spanning tree (as projected into 2D by t-SNE if required). Parameters ---------- axis : matplotlib axis, optional The axis to render the plot to node_size : int, optional The size of nodes in the plot (default 40). node_color : matplotlib color spec, optional The color to render nodes (default black). node_alpha : float, optional The alpha value (between 0 and 1) to render nodes with (default 0.8). edge_cmap : matplotlib colormap, optional The colormap to color edges by (varying color by edge weight/distance). Can be a cmap object or a string recognised by matplotlib. (default `viridis_r`) edge_alpha : float, optional The alpha value (between 0 and 1) to render edges with (default 0.5). edge_linewidth : float, optional The linewidth to use for rendering edges (default 2). vary_line_width : bool, optional Edge width is proportional to (log of) the inverse of the mutual reachability distance. (default True) colorbar : bool, optional Whether to draw a colorbar. (default True) Returns ------- axis : matplotlib axis The axis used the render the plot. rN)LineCollectionzJYou must install the matplotlib library to plot the minimum spanning tree.izDToo many data points for safe rendering of an minimal spanning tree!rr )Z n_componentsr)rnrealpha)rSrsrwzMutual reachability distance)r}r~matplotlib.collectionsrrrr"rrrZ fit_transformrrrrCrrr=astyperrrscatterrZ set_yticksrrxr)r,r node_size node_color node_alpha edge_alpha edge_cmapZedge_linewidthrrrrZdata_for_projectionZ projection line_widthZ line_coordsZline_collectionrrrrrsB.   ,      zMinimumSpanningTree.plotcCs |jS)zLReturn a numpy array of weighted edges in the minimum spanning tree )rrrrrrrYszMinimumSpanningTree.to_numpycCsfzddlm}Wntk r,tdYnX||jjdt|jjdt|jjdd}|S)a"Return a Pandas dataframe of the minimum spanning tree. Each row is an edge in the tree; the columns are `from`, `to`, and `distance` giving the two vertices of the edge which are indices into the dataset, and the distance between those datapoints. rrrrr )fromtor)rrrrrrr)r,rrrrrr^s zMinimumSpanningTree.to_pandascCszddlm}m}Wntk r0tdYnX|}|jD] }|j|d|d|ddq>ddt|jD}|||d |S) a Return a NetworkX Graph object representing the minimum spanning tree. Edge weights in the graph are the distance between the nodes they connect. Nodes have a `data` attribute attached giving the data vector of the associated point. r)Graphrrrr rcSsi|]\}}|t|qSr)tuple)rindexrUrrrrYsz3MinimumSpanningTree.to_networkx..r)rrrrrrrr)r,rrrrUZ data_dictrrrrps  zMinimumSpanningTree.to_networkx) Nrrrrrr TT)rrrr-rrrrrrrrrs \rc@s>eZdZdZdddZdd d Zd d ZddZdddZdS)ApproximationGrapha Cluster approximation graph describing the connectivity in clusters that is used to detect branches. Parameters ---------- approximation_graphs : list[np.ndarray], shape (n_clusters), labels : np.ndarray, shape (n_samples, ) cluster and branches labelling. probabilities : np.ndarray, shape (n_samples, ) cluster and branches probabilities. cluster_labels : np.ndarray, shape (n_samples, ) HDBSCAN* labelling. cluster_probabilities : np.ndarray, shape (n_samples, ) HDBSCAN* probabilities. cluster_centralities : np.ndarray, shape (n_samples, ) Within cluster centrality values. branch_labels : np.ndarray, shape (n_samples, ) Within cluster branch labels for each point. branch_probabilities : np.ndarray, shape (n_samples, ) Within cluster branch membership strengths for each point. Attributes ---------- point_mask : np.ndarray[bool], shape (n_samples) A mask to extract points within clusters from the raw data. Nc stjddt|Ddtjfdtjfdtjfdtjfdtjfgd|_d k|_| dk rj| |jddfnd|_tjfd dt|jd Dd tjfd tjfd tjfdtjfdtjfdtjfdtjfdtjfgd|_ d|_ dS)NcSs8g|]0\}}|D]"}|d|d|d|d|fqqS)rrr r r)rrOedgesedgerrrrsz/ApproximationGraph.__init__..rr centralitymutual_reachabilityrO)dtyperc s<g|]4}||||||||fqSrrrrT branch_labelsbranch_probabilitiescluster_centralitiescluster_labelscluster_probabilitieslabels probabilitiesrrrs idlabelZ probabilityZ cluster_labelZcluster_probabilityZcluster_centralityZ branch_labelZbranch_probability) rrrZintpZfloat64_edges point_mask _raw_datawhere_points_pos) r,Zapproximation_graphsrrrrrrrraw_datarrrr-s8    zApproximationGraph.__init__rrbrorc Cszddlm}ddlm}Wntk r8tdYnX|dkrDnt|tr||jjj krd|krd}d}d}|j|d}q|j|}nn|j dk r|dk r||kr| |}|j dd|f}n6|j dk r| drt |d d}|j dd|f}nt|t|jkr||j}t| tr@| |jjj kr@|j| } tjtt|j|_tjtt|j|_|dkr.z ddl}Wntk rtd YnX|jdkr|}|jD]&}|j|d |d d |ddq|jj|dd|_|jD]&\}}|d|j|<|d |j|<qn|jdt|jkrj|dddf|_|ddd f|_nf|jdt|jkrt|jdD]6\}}||df|j|df<||d f|j|d f<qntd|jd }|jd }|j t!t"t"|j||j|t"|j||j||| |dd}|#| | t| trJ|$| n |%| |dk rh|&||'(||j)|j|j|j|j|d| |ddd|j)|j|j|j|j|||| |dd||d |*ddS)aE Plots the Approximation graph, requires networkx and matplotlib. Parameters ---------- positions : np.ndarray, shape (n_samples, 2) (default = None) A position for each data point in the graph or each data point in the raw data. When None, the function attempts to compute graphviz' sfdp layout, which requires pygraphviz to be installed and available. node_color : str (default = 'label') The point attribute to to color the nodes by. Possible values: - id - label - probability - cluster_label - cluster_probability - cluster_centrality - branch_label - branch_probability, - The input data's feature (if available) names if ``feature_names`` is specified or ``feature_x`` for the x-th feature if no ``feature_names`` are given, or anything matplotlib scatter interprets as a color. node_vmin : float, (default = None) The minimum value to use for normalizing node colors. node_vmax : float, (default = None) The maximum value to use for normalizing node colors. node_cmap : str, (default = 'tab10') The cmap to use for coloring nodes. node_alpha : float, (default = 1) The node transparency value. node_size : float, (default = 5) The node marker size value. node_marker : str, (default = 'o') The node marker string. edge_color : str (default = 'label') The point attribute to to color the nodes by. Possible values: - weight - mutual reachability - centrality, - cluster, or anything matplotlib linecollection interprets as color. edge_vmin : float, (default = None) The minimum value to use for normalizing edge colors. edge_vmax : float, (default = None) The maximum value to use for normalizing edge colors. edge_cmap : str, (default = viridis) The cmap to use for coloring edges. edge_alpha : float, (default = 1) The edge transparency value. edge_width : float, (default = 1) The edge line width size value. rNzHYou must install the matplotlib library to plot the Approximation Graph.r Ztab10rqZfeature_z7You must install the networkx to compute a sfdp layout.rr rrrZsfdp)progrz(Incorrect number of positions specified.)rrerZzordersilverrc)rlmarkerrrnru)rerrrnruZvminZvmaxoff)+r}r~r collectionsrr>rrrnamesrr startswithrrrrrnanZonesZ_xsZ_ysrrrrZ nx_agraphZgraphviz_layoutitemsr"rr_rlistr@Zset_climZ set_edgecolorrZ set_alpharZadd_collectionrr)r,Z positions feature_namesrZ node_vminZ node_vmaxZ node_cmaprrZ node_markerZ edge_colorZ edge_vminZ edge_vmaxrrZ edge_widthrZmcidxnxgrUrvrTdsourcetargetlcrrrrsU                           zApproximationGraph.plotcCs|j|jfS)aConverts the approximation graph to numpy arrays. Returns ------- points : np.recarray, shape (n_points, 8) A numpy record array with for each point its: - id (row index), - label, - probability, - cluster label, - cluster probability, - cluster centrality, - branch label, - branch probability edges : np.recarray, shape (n_edges, 5) A numpy record array with for each edge its: - parent point, - child point, - cluster centrality, - mutual reachability, - cluster label )rrrrrrrrszApproximationGraph.to_numpycCsJzddlm}Wntk r,tdYnX||j}||j}||fS)aConverts the approximation graph to pandas data frames. Returns ------- points : pd.DataFrame, shape (n_points, 8) A DataFrame with for each point its: - id (row index), - label, - probability, - cluster label, - cluster probability, - cluster centrality, - branch label, - branch probability edges : pd.DataFrame, shape (n_edges, 5) A DataFrame with for each edge its: - parent point, - child point, - cluster centrality, - mutual reachability, - cluster label rrr)rrrrr)r,rZpointsrrrrrs   zApproximationGraph.to_pandasc Csz ddl}Wntk r(tdYnX|}|jD]6}|j|d|dd|d|d|d|d d q8|jjjddD] }||t |jd |g|q|j dk r|dkrd d t |j j dD}t |D]4\}}||t t|jd |j dd|f|q|S)aConvert to a NetworkX Graph object. Parameters ---------- feature_names : list[n_features] Names to use for the data features if available. Returns ------- g : nx.Graph A NetworkX Graph object containing the non-noise points and edges within clusters. Node attributes: - label, - probability, - cluster label, - cluster probability, - cluster centrality, - branch label, - branch probability, Edge attributes: - weight (1 / mutual_reachability), - mutual_reachability, - centrality, - cluster label, - rNrrr rrrrO)rrrrOrcSsg|]}d|qS)zfeature rrrrrr sz2ApproximationGraph.to_networkx..)rrrrrrrrrr?rrAr"rr@)r,rr r rUattrr namerrrrs8       zApproximationGraph.to_networkx)N)NNrNNrbrrrrNNrbrr)N) rrrrr-rrrrrrrrrs.- 9 J#r)numpyrZscipy.cluster.hierarchyrZsklearn.manifoldrZsklearn.decompositionrZsklearn.metricsrwarningsrZ _hdbscan_treerr r rErFrGrHrr r%objectr&rrrrrrrrrs0      La