U Xe@snddlZddlZddlmZzddlZddlZddl m Z ddl m ZddlmZddlZddlZddlZddlmZddlmZddlZddlmmZ Wn&e!k rede!ddYnXddl"Z#ddl$Z#ddl%Z#ddl&m'Z'ddl(m)Z)m*Z*ddlm+Z,ddlm-Z-m.Z.dd l/m0Z0dd l1m2Z2m3Z3ddlm+Z4ddlmZej5j67d ej8Z9ej5j67d ej:Z;ej5j67d ej<Z=ej5j6j7dej>ddddZ?ej5j67dej@ZAejBjCe9d dejBjCe;d dejBjCe=d dejBjCe?ddejBjCeAddddZDeEddgddZFeEddgddZGeEddgddZHd ddd dd d!dd"dd#d!dd"dd$d%d&d'dd(d)d&d'dd*d+d&d'dd dddddddd dd ddddd, ZIeJd-d.d/d0d1gZKd2d3ZLd4d5ZMd6d7ZNd8d9ZOd:d;ZPejQdd?ZRd@dAZSdBdCZTd[dGdHZUd\dIdJZVdKdLZ+d]dMdNZWd^dOdPZXd_dSdTZYd`dVdWZZdadYdZZ[dS)bN)warnaThe umap.plot package requires extra plotting libraries to be installed. You can install these via pip using pip install umap-learn[plot] or via conda using conda install pandas matplotlib datashader bokeh holoviews colorcet scikit-image ziumap.plot requires pandas matplotlib datashader bokeh holoviews scikit-image and colorcet to be installedPatch) submatrixaverage_nn_distance)show) output_fileoutput_notebook)column)CustomJS TextInputfiredarkblue darkgreendarkred)colorsNZ darkpurple)namecCsdd|DS)NcSsg|]}tj|qS matplotlibrZto_hex.0crr(lib/python3.8/site-packages/umap/plot.py Jsz_to_hex..r)Zarrrrr_to_hexIsrz uint8(uint32)cCs |d@d?S)Nirxrrr_redMsr"cCs |d@d?S)Nirr rrr_greenRsr$cCs|d@S)Nrr rrr_blueWsr&Zrainbowblack)cmapcolor_key_cmap background edge_cmapviridisSpectralZgrayinfernoBluesZtab20whitegray_rZRedsZtab20bZGreensZtab20c) r r,r.ZblueZredZgreenrrrpcaicavq local_dim neighborhoodcCs,t|dr|jSt|dr |jStddS)N embedding_ embeddingz1Could not find embedding attribute of umap_object)hasattrr7r8 ValueError umap_objectrrr_get_embeddings   r=cCst|dr|jSdSdS)Nmetric euclidean)r9r>r;rrr _get_metrics r@cCst|dr|jSiSdS)N _metric_kwds)r9rAr;rrr_get_metric_kwdss rBcCs:|jddd}tt|t|t|g}|||S)N)datanpZdstackr&r$r"Zimshow)Zdatashader_imageaxZimg_revZmpl_imgrrr_embed_datashader_in_an_axiss rGcCst|drj|jrjtj|j}t||ddd|f}t|||}t |}t|||}t|||}n&tj dtj d}|j j |j|d\}}||fS)N _small_dataZdtypek)r9rHsklearnmetricspairwise_distances _raw_datarEZ argpartitionrZargsortemptyint64Z_knn_search_indexquery)r< nhood_sizeZdmatindicesZdmat_shortenedZindices_sortedZdistsZ rng_staterrr _nhood_searchs    rVF)ZnopythoncCs|t|jd}t|jdD]X}tj||||ddjd}tt||||gjd}t|t|||<q|S)z*Compute Jaccard index of two neighborhoodsrT)Z assume_unique)rErQshaperangeZ intersect1duniqueZhstackfloat)Z indices_leftZ indices_rightresultiZintersection_sizeZ union_sizerrr_nhood_compares"r]c Cst|dddf}t|dddf}t|dddf}t|dddf}t|d||t|d||t|d||t|d||f}|S)z2Compute bounds on a space with appropriate paddingNrg?)rEZnanminZnanmaxround)pointsZmin_xZmax_xZmin_yZmax_yextentrrr _get_extentsrbcCsf|dkrd}nT|dr^tdd|dd|dd|dd fD}|d krXd}qbd}nd}|S) Nr'r0#cSsg|]}td|qS)Z0x)intrrrrrsz&_select_font_color..r^rI~) startswithrEZmean)r* font_colorZmean_valrrr_select_font_colors *rj Tr%c s"t|} tj|| | d| df| d| dfd} tj|dd}d}|dk rV|jd|jdkr~td |jd|jdt||d <| j |d d t d d }dkr|dkrt j |d| d}ndkr*t |}|jd}tt|t dd|fddt|D}t j |d| d}n(fddD}t j |d| d}nz|dk r|jd|jdkrtd|jd|jdt |}|jddkr:t |t |}}||d}tt |||t j|d<| j |d d t dd }tt|t dddt j |d| d}nht||d<| j |d d t dd }tt|t dd|jd}tt||t j |d| d}n,| j |d d td }t j |t|| d}|dk rt ||}|dk rt||| r|dk r|j|d|S|SdS)zUse datashader to plot pointsrr^rIZ plot_widthZ plot_heightZx_rangeZy_ranger!ycolumnsN?Labels must have a label for each sample (size mismatch: {} {})labelr!roZaggeq_hist)howalphacs g|]\}}t||dqS)Z facecolorrsrrr\rL color_keyrrr+sz%_datashade_points..)r{rvrwcsg|]}t||dqSrxrrrLrzrrr3s?Values must have a value for each sample (size mismatch: {} {})rgo@val_cat)r(rwZhandles) rbdsCanvaspd DataFramerWr:format Categoricalr` count_cattfshaderErYrpltget_cmaplinspace enumeratekeysminmaxr_astypeZint16dictzipcountset_backgroundrGlegend)r`rFlabelsvaluesr(r{r)r*widthheight show_legendrwracanvasrDlegend_elementsZ aggregationr[ unique_labels num_labelsZ unique_valuesmin_valZmax_valZbin_sizeZcolor_key_colsrrzr_datashade_pointss               rc sjdt|jd} d} |dkrLtjd}tj||| |fd}|d}|||dk r|jd|jdkrtd |jd|jddkrt |jd}t |t dd|fd d t D} ttrt|}t |fd d D} nbt |tjdkr.csg|]}t||dqSrxrr|rzrrrs:Color key must have enough colors for the number of labelscs"i|]\}}|tj|qSrrryrzrr sz&_matplotlib_points..cs g|]\}}t||dqSrxrryrzrrrssrrwr})rrr(rw?)rrr)rEsqrtrWrrcParamsfigure add_subplotZ set_facecolorr:rrYrrr isinstancerrSeriesmaplenscatterr)r`rFrrr(r{r)r*rrrrw point_sizerdpifigrr new_color_keycolorrrr_matplotlib_pointsks|             . * rcCsZt|tjrtnBt|tjr*t|n,t|tjj j rNtt |ddnt ddS)zDisplay a plot, either interactive or static. Parameters ---------- plot_to_show: Output of a plotting command (matplotlib axis or bokeh figure) The plot to show Returns ------- None bokeh)Zbackendz>The type of ``plot_to_show`` was not valid, or not understood.N) rrZAxes show_staticbplrshow_interactivehvcoreZspacesZ DynamicMapZrenderr:)Z plot_to_showrrrrs   rc Cs|dk r,t|d}t|d}t|d}|dk rD|dk rDtd|dk rjd|kr`dksjntd|dkrzt|}| dk rt| |jd krtd t| |jd || }|dk r|| }|dk r|| }|jd d krtd t|}| dkr*tjd}tj | || |fd}| d} |jd | | dkrbt || ||||||| | | | } n6|dk rv|d}nd}t || ||||||| | | | } | j ggdt|dkr| jdddt||j|j| jd|dn$| jddd|j|j| jd|d| S)aPlot an embedding as points. Currently this only works for 2D embeddings. While there are many optional parameters to further control and tailor the plotting, you need only pass in the trained/fit umap model to get results. This plot utility will attempt to do the hard work of avoiding over-plotting issues, and make it easy to automatically colour points by a categorical labelling or numeric values. This method is intended to be used within a Jupyter notebook with ``%matplotlib inline``. Parameters ---------- umap_object: trained UMAP object A trained UMAP object that has a 2D embedding. points: array, shape (n_samples, dim) (optional, default None) An array of points to be plotted. Usually this is None and so the original embedding points of the umap_object are used. However points can be passed explicitly instead which is useful for points manually transformed. labels: array, shape (n_samples,) (optional, default None) An array of labels (assumed integer or categorical), one for each data sample. This will be used for coloring the points in the plot according to their label. Note that this option is mutually exclusive to the ``values`` option. values: array, shape (n_samples,) (optional, default None) An array of values (assumed float or continuous), one for each sample. This will be used for coloring the points in the plot according to a colorscale associated to the total range of values. Note that this option is mutually exclusive to the ``labels`` option. theme: string (optional, default None) A color theme to use for plotting. A small set of predefined themes are provided which have relatively good aesthetics. Available themes are: * 'blue' * 'red' * 'green' * 'inferno' * 'fire' * 'viridis' * 'darkblue' * 'darkred' * 'darkgreen' cmap: string (optional, default 'Blues') The name of a matplotlib colormap to use for coloring or shading points. If no labels or values are passed this will be used for shading points according to density (largely only of relevance for very large datasets). If values are passed this will be used for shading according the value. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. color_key: dict or array, shape (n_categories) (optional, default None) A way to assign colors to categoricals. This can either be an explicit dict mapping labels to colors (as strings of form '#RRGGBB'), or an array like object providing one color for each distinct category being provided in ``labels``. Either way this mapping will be used to color points according to the label. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. color_key_cmap: string (optional, default 'Spectral') The name of a matplotlib colormap to use for categorical coloring. If an explicit ``color_key`` is not given a color mapping for categories can be generated from the label list and selecting a matching list of colors from the given colormap. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. background: string (optional, default 'white) The color of the background. Usually this will be either 'white' or 'black', but any color name will work. Ideally one wants to match this appropriately to the colors being used for points etc. This is one of the things that themes handle for you. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. width: int (optional, default 800) The desired width of the plot in pixels. height: int (optional, default 800) The desired height of the plot in pixels show_legend: bool (optional, default True) Whether to display a legend of the labels subset_points: array, shape (n_samples,) (optional, default None) A way to select a subset of points based on an array of boolean values. ax: matplotlib axis (optional, default None) The matplotlib axis to draw the plot to, or if None, which is the default, a new axis will be created and returned. alpha: float (optional, default: None) The alpha blending value, between 0 (transparent) and 1 (opaque). Returns ------- result: matplotlib axis The result is a matplotlib axis with the relevant plot displayed. If you are using a notebooks and have ``%matplotlib inline`` set then this will simply display inline. Nr(r)r*?Conflicting options; only one of labels or values should be set?'Alpha must be between 0 and 1 inclusiverESize of subset points ({}) does not match number of input points ({})r^rl8Plotting is currently only implemented for 2D embeddingsrrr r%ZxticksZyticksr?Gz?{Gz?z,UMAP: metric={}, n_neighbors={}, min_dist={}right transformZhorizontalalignmentr!UMAP: n_neighbors={}, min_dist={})_themesr:r=rrWrrjrrrrrrsetr@text n_neighborsmin_dist transAxes)r<r`rrthemer(r{r)r*rrr subset_pointsrFrwrirrrrrr`s           r`c  CsX|dk r8t|d}t|d} t|d}t|d} t|} tj| dd}dt| jd }|d krztt|}nd }|rd }nd }|j }tjt |j |j |jgjd d}|jtj|d<|jtj|d<t| }tj| | |d |d f|d|dfd}|dkr*tj||dd}n2|dkrNtdtj||dd}ntd|tj|j|ddt ddt!"||d}t#|| }|rt$| d||||| d| | d }|d krtj%|d|d}tj&||dd }n|}t'| }t!j(d!}t!j)| || |fd"}|*d#}t+|||j,ggd$|j-d%d&d'|j.|j/|j0d(|d)|S)*aPlot connectivity relationships of the underlying UMAP simplicial set data structure. Internally UMAP will make use of what can be viewed as a weighted graph. This graph can be plotted using the layout provided by UMAP as a potential diagnostic view of the embedding. Currently this only works for 2D embeddings. While there are many optional parameters to further control and tailor the plotting, you need only pass in the trained/fit umap model to get results. This plot utility will attempt to do the hard work of avoiding over-plotting issues and provide options for plotting the points as well as using edge bundling for graph visualization. Parameters ---------- umap_object: trained UMAP object A trained UMAP object that has a 2D embedding. edge_bundling: string or None (optional, default None) The edge bundling method to use. Currently supported are None or 'hammer'. See the datashader docs on graph visualization for more details. edge_cmap: string (default 'gray_r') The name of a matplotlib colormap to use for shading/ coloring the edges of the connectivity graph. Note that the ``theme``, if specified, will override this. show_points: bool (optional False) Whether to display the points over top of the edge connectivity. Further options allow for coloring/ shading the points accordingly. labels: array, shape (n_samples,) (optional, default None) An array of labels (assumed integer or categorical), one for each data sample. This will be used for coloring the points in the plot according to their label. Note that this option is mutually exclusive to the ``values`` option. values: array, shape (n_samples,) (optional, default None) An array of values (assumed float or continuous), one for each sample. This will be used for coloring the points in the plot according to a colorscale associated to the total range of values. Note that this option is mutually exclusive to the ``labels`` option. theme: string (optional, default None) A color theme to use for plotting. A small set of predefined themes are provided which have relatively good aesthetics. Available themes are: * 'blue' * 'red' * 'green' * 'inferno' * 'fire' * 'viridis' * 'darkblue' * 'darkred' * 'darkgreen' cmap: string (optional, default 'Blues') The name of a matplotlib colormap to use for coloring or shading points. If no labels or values are passed this will be used for shading points according to density (largely only of relevance for very large datasets). If values are passed this will be used for shading according the value. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. color_key: dict or array, shape (n_categories) (optional, default None) A way to assign colors to categoricals. This can either be an explicit dict mapping labels to colors (as strings of form '#RRGGBB'), or an array like object providing one color for each distinct category being provided in ``labels``. Either way this mapping will be used to color points according to the label. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. color_key_cmap: string (optional, default 'Spectral') The name of a matplotlib colormap to use for categorical coloring. If an explicit ``color_key`` is not given a color mapping for categories can be generated from the label list and selecting a matching list of colors from the given colormap. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. background: string (optional, default 'white) The color of the background. Usually this will be either 'white' or 'black', but any color name will work. Ideally one wants to match this appropriately to the colors being used for points etc. This is one of the things that themes handle for you. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. width: int (optional, default 800) The desired width of the plot in pixels. height: int (optional, default 800) The desired height of the plot in pixels Returns ------- result: matplotlib axis The result is a matplotlib axis with the relevant plot displayed. If you are using a notebook and have ``%matplotlib inline`` set then this will simply display inline. Nr(r)r+r*rnrprrr^logru)sourcetargetweightrrrlrIrmr)rZhammerzYHammer edge bundling is expensive for large graphs! This may take a long time to compute!z&{} is not a recognised bundling methodr!rort)r(rvFr)Z thresholdZmax_pxZover)rvrrrrrrrrr)1rr=rrrErrWrdr_graph_ZtocooZvstackrowcolrDTrrint32rrbrrbdZdirectly_connect_edgesrZ hammer_bundler:rrrlinesumrrrrZ dynspreadstackrjrrrrGrrrrr)r<Z edge_bundlingr+Z show_pointsrrrr(r{r)r*rrr`Zpoint_dfrZpx_sizeZedge_howZ coo_graphZedge_dfrarZedgesZedge_imgZ point_imgr[rirrrFrrr connectivitys             r皙?c " Cst|} | jddkrtd|dkr:dt| jd}|dkr\tjd} |dkr\|d 9}t|} |dkr|d krt} | d }|d kr2t j j d d |j}|t|8}|tj|dd}|j| dddf| dddf||dd|d|jddd|j|j|jd| d|jggdnn|dkrt j jd d |j}|t|8}|tj|dd}|j| dddf| dddf||dd|d|jddd|j|j|jd| d|jggdn|dkrt jjd d|j}t j|j|j}|t|8}|tj|dd}|j| dddf| dddf||dd|d|jddd|j|j|jd| d|jggdn|dkrt ||\}}t j!"| }|j#| |d\}}t$|%tj&|%tj&}t'|d }t'|d!}|j| dddf| dddf|||||d"|d#|jddd|j|j|jd| d|jggdt(j)j*||d$}t(j+j,||d%}|-|tj.||d&n|d'krt ||j\}}|j}tj/|jdtj0d(}t1|jdD]@}t j |||}t2t3|j4|kdd||<qt'|d }t'|d!}|j| dddf| dddf|||||d"|d)|jddd|j|j|jd| d|jggdt(j)j*||d$}t(j+j,||d%}|-|tj.||d&n|d krt5t6t7d*d}t6t7|d}tj8||d+d,d-\} } | j9} | t6t7dD]}|:qJt;| t7D]\}}!t<||!||d.d/qdntd0d1=t>t7d2|S)3aJ Provide a diagnostic plot or plots for a UMAP embedding. There are a number of plots that can be helpful for diagnostic purposes in understanding your embedding. Currently these are restricted to methods of coloring a scatterplot of the embedding to show more about how the embedding is representing the data. The first class of such plots uses a linear method that preserves global structure well to embed the data into three dimensions, and then interprets such coordinates as a color space -- coloring the points by their location in the linear global structure preserving embedding. In such plots one should look for discontinuities of colour, and consider overall global gradients of color. The diagnostic types here are ``'pca'``, ``'ica'``, and ``'vq'`` (vector quantization). The second class consider the local neighbor structure. One can either look at how well the neighbor structure is preserved, or how the estimated local dimension of the data varies. Both of these are available, although the local dimension estimation is the preferred option. You can access these are diagnostic types ``'local_dim'`` and ``'neighborhood'``. Finally the diagnostic type ``'all'`` will provide a grid of diagnostic plots. Parameters ---------- umap_object: trained UMAP object A trained UMAP object that has a 2D embedding. diagnostic_type: str (optional, default 'pca') The type of diagnostic plot to show. The options are * 'pca' * 'ica' * 'vq' * 'local_dim' * 'neighborhood' * 'all' nhood_size: int (optional, default 15) The size of neighborhood to compare for local neighborhood preservation estimates. local_variance_threshold: float (optional, default 0.8) To estimate the local dimension we consider a PCA of the local neighborhood and estimate the dimension as that which provides ``local_variance_threshold`` or more of the ``variance_explained_ratio``. ax: matplotlib axis (optional, default None) A matplotlib axis to plot to, or, if None, a new axis will be created and returned. cmap: str (optional, default 'viridis') The name of a matplotlib colormap to use for coloring points if the ``'local_dim'`` or ``'neighborhood'`` option are selected. point_size: int (optional, None) If provided this will fix the point size for the plot(s). If None then a suitable point size will be estimated from the data. Returns ------- result: matplotlib axis The result is a matplotlib axis with the relevant plot displayed. If you are using a notebook and have ``%matplotlib inline`` set then this will simply display inline. r^rlrNrrr)r5r6g?allrr2rI)Z n_components)axisgQ?rz&Colored by RGB coords of PCA embeddingrrrrrrr3z*Colored by RGB coords of FastICA embeddingr4)Z n_clustersz,Colored by RGB coords of Vector Quantizationr6rKre_)rrr(vminvmaxz%Colored by neighborhood Jaccard index)rr)normr()rFr5rJz!Colored by approx local dimensionr)rrT)rZconstrained_layoutg@)diagnostic_typerFrz%Unknown diagnostic; should be one of z, z or "all")?r=rWr:rErrrrjrrrMZ decompositionZPCAZ fit_transformrPrrrZ set_titlerrrrrrZFastICAZclusterZKMeansZfitrNrOZcluster_centers_rVZ neighborsZKDTreerSr]rrZ percentilerrZ NormalizecmZScalarMappableZ set_arrayZcolorbarrQrRrXwhereZcumsumZexplained_variance_ratio_rdr_diagnostic_typesZsubplotsZflatremover diagnosticjoinlist)"r<rrTZlocal_variance_thresholdrFr(rr*rrr`rrirZ color_projZcolor_projectorZ highd_indicesZ highd_distsZtreeZ lowd_distsZ lowd_indicesZaccuracyrrrZmappablerDr5r\r2ZcolsZrowsZaxsZplt_typerrrrs`S   ,   ,   ,                     rffffff?c%s6|dk r,t|d}t|d}t|d} |dk rD|dk rDtd|dk rjd|kr`dksjntdt|}| dk rt| |jd krtd t| |jd || }|jd d krtd | dkrdt|jd } tj t|dd}|dk rt ||d<dkrDt |}|jd }t t |td d |ttrft||d<nPt |}t|jd krtdfddt|D}t|||d<d}nj|dk rt ||d<t t |td d d}tjd|t|t|d}ntjt |d}| dk rJ|| }|dk rJ|| }|jd | | dkrd}d}|dk ri}|D]"}||||<d|d||<q|t|}|dk r|D]}|jjdkrd}qڐq|dk r||d <nd |d <t !|}t j"| | |sdn||dk r"|nd!| d"}|j#d#d$||| d d%d|j$_%d|j&_%|r2t'd&d'd(} |dkrg}|dk r|(|j)|dk r|*dt|d krt+d)n2t,t||d ||d*d+d,}!| -d|!t.| |}nH|dk rt+d-|r t+d.|dk rt+d/t/0d0t/j1d1d2t/j23t/j2j4| ddd3|dk rt/j5|d#d$gd4}"t6j7|"t89dt || | d5}n|dk r|j:}#|j:|#}$t;|j:|#|$d|d6<t/j5|d#d$gd6gd7}"t6j7|"t89d6t || | d8}n0t/j5|d#d$gd4}"t6j7|"t8<t || | d8}|S)9aCreate an interactive bokeh plot of a UMAP embedding. While static plots are useful, sometimes a plot that supports interactive zooming, and hover tooltips for individual points is much more desirable. This function provides a simple interface for creating such plots. The result is a bokeh plot that will be displayed in a notebook. Note that more complex tooltips etc. will require custom code -- this is merely meant to provide fast and easy access to interactive plotting. Parameters ---------- umap_object: trained UMAP object A trained UMAP object that has a 2D embedding. labels: array, shape (n_samples,) (optional, default None) An array of labels (assumed integer or categorical), one for each data sample. This will be used for coloring the points in the plot according to their label. Note that this option is mutually exclusive to the ``values`` option. values: array, shape (n_samples,) (optional, default None) An array of values (assumed float or continuous), one for each sample. This will be used for coloring the points in the plot according to a colorscale associated to the total range of values. Note that this option is mutually exclusive to the ``labels`` option. hover_data: DataFrame, shape (n_samples, n_tooltip_features) (optional, default None) A dataframe of tooltip data. Each column of the dataframe should be a Series of length ``n_samples`` providing a value for each data point. Column names will be used for identifying information within the tooltip. tools: List (optional, default None), Defines the tools to be configured for interactive plots. The list can be mixed type of string and tools objects defined by Bokeh like HoverTool. Default tool list Bokeh uses is ["pan","wheel_zoom","box_zoom","save","reset","help",]. When tools are specified, and includes hovertool, automatic tooltip based on hover_data is not created. theme: string (optional, default None) A color theme to use for plotting. A small set of predefined themes are provided which have relatively good aesthetics. Available themes are: * 'blue' * 'red' * 'green' * 'inferno' * 'fire' * 'viridis' * 'darkblue' * 'darkred' * 'darkgreen' cmap: string (optional, default 'Blues') The name of a matplotlib colormap to use for coloring or shading points. If no labels or values are passed this will be used for shading points according to density (largely only of relevance for very large datasets). If values are passed this will be used for shading according the value. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. color_key: dict or array, shape (n_categories) (optional, default None) A way to assign colors to categoricals. This can either be an explicit dict mapping labels to colors (as strings of form '#RRGGBB'), or an array like object providing one color for each distinct category being provided in ``labels``. Either way this mapping will be used to color points according to the label. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. color_key_cmap: string (optional, default 'Spectral') The name of a matplotlib colormap to use for categorical coloring. If an explicit ``color_key`` is not given a color mapping for categories can be generated from the label list and selecting a matching list of colors from the given colormap. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. background: string (optional, default 'white') The color of the background. Usually this will be either 'white' or 'black', but any color name will work. Ideally one wants to match this appropriately to the colors being used for points etc. This is one of the things that themes handle for you. Note that if theme is passed then this value will be overridden by the corresponding option of the theme. width: int (optional, default 800) The desired width of the plot in pixels. height: int (optional, default 800) The desired height of the plot in pixels point_size: int (optional, default None) The size of each point marker subset_points: array, shape (n_samples,) (optional, default None) A way to select a subset of points based on an array of boolean values. interactive_text_search: bool (optional, default False) Whether to include a text search widget above the interactive plot interactive_text_search_columns: list (optional, default None) Columns of data source to search. Searches labels and hover_data by default. interactive_text_search_alpha_contrast: float (optional, default 0.95) Alpha value for points matching text search. Alpha value for points not matching text search will be 1 - interactive_text_search_alpha_contrast alpha: float (optional, default: None) The alpha blending value, between 0 (transparent) and 1 (opaque). Returns ------- Nr(r)r*rrrrrrr^rlrrrnrprsrrcsi|]\}}||qSrrryrzrrrszinteractive..valuer)ZlowZhighrrTz@{}Z HoverToolFrwz'pan,wheel_zoom,box_zoom,save,reset,help)rrtooltipstoolsZbackground_fill_colorr!ro)r!rorrsizerwzSearch:)rtitlezinteractive_text_search_columns set to True, but no hover_data or labels provided.Please provide hover_data or labels to use interactive text search.)rZmatching_alphaZnon_matching_alphaZsearch_columnsa var data = source.data; var text_search = cb_obj.value; var search_columns_dict = {} for (var col in search_columns){ search_columns_dict[col] = search_columns[col] } // Loop over columns and values // If there is no match for any column for a given row, change the alpha value var string_match = false; for (var i = 0; i < data.x.length; i++) { string_match = false for (var j in search_columns_dict) { if (String(data[search_columns_dict[j]][i]).includes(text_search) ) { string_match = true } } if (string_match){ data['alpha'][i] = matching_alpha }else{ data['alpha'][i] = non_matching_alpha } } source.change.emit(); )argscodezbToo many points for hover data -- tooltips will notbe displayed. Sorry; try subsampling your data.zAToo many points for text search.Sorry; try subsampling your data.z6Alpha parameter will not be applied on holoviews plotsri,)r)ZbgcolorZxaxisZyaxis)kdims) aggregatorr{r(rrr~)rZvdims)rr(rr)=rr:r=rrWrrErrrZasarrayrYrrrrrrrrrbtrZ linear_cmaprrrrZrgb2hexritems __class____name__rZColumnDataSourcerZcircleZgridZvisiblerr extendrqappendrr Z js_on_changer r extensionoutputZoptsdefaultsZRGBZPointshdZ datashaderrrrr)%r<rrZ hover_datarrr(r{r)r*rrrrZinteractive_text_searchZinteractive_text_search_columnsZ&interactive_text_search_alpha_contrastrwr`rDrrrrZpaletterZtooltip_neededZ tooltip_dictZcol_nameZ_toolZ data_sourceZplotZ text_inputcallbackZ point_plotrZ val_rangerrzr interactivesF                            $        r cCsJt|j}|dkr$t}|d}|d|d|j||d|S)aCreate a histogram of the average distance to each points nearest neighbors. Parameters ---------- umap_object: trained UMAP object A trained UMAP object that has an embedding. bins: int (optional, default 25) Number of bins to put the points into ax: matplotlib axis (optional, default None) A matplotlib axis to plot to, or, if None, a new axis will be created and returned. Returns ------- Nrz%Average distance to nearest neighborsZ Frequency)bins)rrrrrZ set_xlabelZ set_ylabelZhist)r<rrFZ nn_distancesrrrrnearest_neighbour_distribution:s    r) NNNr/Nr-r0rkrkTr%) NNNr/Nr-r0rkrkTN)NNNNr/Nr-r0rkrkTNNN) Nr1FNNNr/Nr-r0rkrk) r2rrNr,Nr0rkrk)NNNNNr/Nr-r0rkrkNNFNrN)r N)\ZnumpyrEZnumbawarningsrZpandasrZ datashaderrZdatashader.transfer_functionsZtransfer_functionsrZdatashader.bundlingZbundlingrZmatplotlib.pyplotZpyplotrZcolorcetZmatplotlib.colorsrZ matplotlib.cmZbokeh.plottingZplottingrZbokeh.transformrrZ holoviewsrZholoviews.operation.datashaderZ operationr  ImportErrorZsklearn.decompositionrMZsklearn.clusterZsklearn.neighborsZmatplotlib.patchesrZ umap.utilsrrrrrr Z bokeh.layoutsr Z bokeh.modelsr r rrZLinearSegmentedColormap from_listr Z fire_cmapZkbcZ darkblue_cmapZkgyZdarkgreen_cmapZlinear_kry_5_95_c72Z darkred_cmapZlinear_bmw_5_95_c89Zdarkpurple_cmapZ colormapsregisterrZ vectorizer"r$r&rZarrayrr=r@rBrGrVZjitr]rbrjrrr`rrr rrrrrs                   9    q ^ x e  y