Return x, y coordinates from Bokeh graph

Question

I'm trying to adapt the answer from: Get selected data contained within box select tool in Bokeh

but get:

NameError: name 'inds' is not defined 

after selecting the points.

Does anyone know whats going on?

Code I am using :

    import pandas as pd
    import numpy as np
    import matplotlib as mpl
    from bokeh.transform import factor_cmap, linear_cmap
    from bokeh.palettes import Spectral6
    from bokeh.plotting import figure, output_file, show
    from bokeh.models import CustomJS, ColumnDataSource
    from bokeh.io import output_notebook

    import bokeh.plotting as bpl
    import bokeh.models as bmo
    from bokeh.models.tools import *

    output_notebook()

    algorithm = 'Loc'
    metric = 'LengthOfStayDaysNBR'
    coordinate_df = pd.DataFrame({'x-{}'.format(str(algorithm)[:3]):np.random.rand(500)\
                                  ,'y-{}'.format(str(algorithm)[:3]):np.random.rand(500)\
                                 ,metric:np.random.rand(500)})


    TOOLS="pan,wheel_zoom,reset,hover,poly_select,lasso_select"


    p = figure(title = '{}'.format(str(algorithm)[:3]),tools=TOOLS)


    source = bpl.ColumnDataSource(coordinate_df)

    if metric == 'LengthOfStayDaysNBR':
        color_mapper = linear_cmap(metric,palette=Spectral6,low=coordinate_df[metric].min(),high=coordinate_df[metric].max())
    else:
        color_mapper = factor_cmap(metric,palette=Spectral6,factors=coordinate_df[metric].unique())



    p1 = figure(width=250, height = 250, title = '{}'.format(str(algorithm)[:3]),tools=TOOLS)

    p1.scatter('x-{}'.format(str(algorithm)[:3]),'y-{}'.format(str(algorithm)[:3]),fill_color= color_mapper, source = source)


    source.callback = CustomJS(args=dict(source=source), code="""
            var inds = cb_obj.get('selected')['1d'].indices;
            var d1 = cb_obj.get('data');
            console.log(d1)
            var kernel = IPython.notebook.kernel;
            IPython.notebook.kernel.execute("inds = " + inds);
            """
    )

    show(p1)


    # Run this after selecting
    for x,y in  zip([source.data['x-{}'.format(str(algorithm)[:3])][i] for i in inds],
        [source.data['y-{}'.format(str(algorithm)[:3])][i] for i in inds]):
        print(x,y)

To give some background: I am trying to take x/y coordinates from a larger dataframe containing many features, select a sub-cohort of the scatterplot, and then graph (in a barchart) the largest (20 or so) averages of the features for the specified x/y coordinates of the sub-cohort. For example, say you have 50 features (columns) and a sub-cohort of 20 data points. I would want the remaining 48 features' averages to be plotted a barchart with the bars' heights representing the features average. Any idea how to do this? – user123328 10 hours ago
Furthermore: it would be great to be able to get the actual indexes/ xy coordinates back into some object, i.e. a dataframe, np.array, list or whatever – user123328 10 hours ago

EDIT:

I got it (kinda) working with the following:

    def app(doc):

        x = df_algs['x-{}'.format(str(algorithm)[:3])]
        y = df_algs['y-{}'.format(str(algorithm)[:3])]
        # create the scatter plot


        if metric == 'LengthOfStayDaysNBR':
            color_mapper = linear_cmap(metric,palette=Spectral6,low=df_algs[metric].min(),high=df_algs[metric].max())
        else:
            color_mapper = factor_cmap(metric,palette=Spectral6,factors=df_algs[metric].unique())


        source = ColumnDataSource(dict(
            x = x
        ,   y = y
        ,   metric_ = df_algs[metric]))
        # create the scatter plot
        p = figure(tools=TOOLS, plot_width=600, plot_height=600, min_border=10, min_border_left=50,
                   toolbar_location="above", x_axis_location=None, y_axis_location=None,
                   title="Linked Histograms")
        p.select(BoxSelectTool).select_every_mousemove = False
        p.select(LassoSelectTool).select_every_mousemove = False

        r = p.scatter('x', 'y', source=source, fill_color = color_mapper, alpha=0.6)
    # 
        # create the horizontal histogram
        hhist, hedges = np.histogram(x, bins=20)
        hzeros = np.zeros(len(hedges)-1)
        hmax = max(hhist)*1.1

        LINE_ARGS = dict(color="#3A5785", line_color=None)

        ph = figure(toolbar_location=None, plot_width=p.plot_width, plot_height=200, x_range=p.x_range,
                    y_range=(-hmax, hmax), min_border=10, min_border_left=50, y_axis_location="right")
        ph.xgrid.grid_line_color = None
        ph.yaxis.major_label_orientation = np.pi/4
        ph.background_fill_color = "#fafafa"

        ph.quad(bottom=0, left=hedges[:-1], right=hedges[1:], top=hhist, color="white", line_color="#3A5785")
        hh1 = ph.quad(bottom=0, left=hedges[:-1], right=hedges[1:], top=hzeros, alpha=0.5, **LINE_ARGS)
        hh2 = ph.quad(bottom=0, left=hedges[:-1], right=hedges[1:], top=hzeros, alpha=0.1, **LINE_ARGS)

        # create the vertical histogram
        vhist, vedges = np.histogram(y, bins=20)
        vzeros = np.zeros(len(vedges)-1)
        vmax = max(vhist)*1.1

        pv = figure(toolbar_location=None, plot_width=200, plot_height=p.plot_height, x_range=(-vmax, vmax),
                    y_range=p.y_range, min_border=10, y_axis_location="right")
        pv.ygrid.grid_line_color = None
        pv.xaxis.major_label_orientation = np.pi/4
        pv.background_fill_color = "#fafafa"

        pv.quad(left=0, bottom=vedges[:-1], top=vedges[1:], right=vhist, color="white", line_color="#3A5785")
        vh1 = pv.quad(left=0, bottom=vedges[:-1], top=vedges[1:], right=vzeros, alpha=0.5, **LINE_ARGS)
        vh2 = pv.quad(left=0, bottom=vedges[:-1], top=vedges[1:], right=vzeros, alpha=0.1, **LINE_ARGS)

        layout = column(row(p, pv), row(ph, Spacer(width=200, height=200)))

        doc.add_root(layout)
        doc.title = "Selection Histogram"

        def update(attr, old, new):
            inds = np.array(new['1d']['indices'])
            if len(inds) == 0 or len(inds) == len(x):
                hhist1, hhist2 = hzeros, hzeros
                vhist1, vhist2 = vzeros, vzeros
            else:
                neg_inds = np.ones_like(x, dtype=np.bool)
                neg_inds[inds] = False
                hhist1, _ = np.histogram(x[inds], bins=hedges)
                vhist1, _ = np.histogram(y[inds], bins=vedges)
                hhist2, _ = np.histogram(x[neg_inds], bins=hedges)
                vhist2, _ = np.histogram(y[neg_inds], bins=vedges)

            hh1.data_source.data["top"]   =  hhist1
            hh2.data_source.data["top"]   = -hhist2
            vh1.data_source.data["right"] =  vhist1
            vh2.data_source.data["right"] = -vhist2

            df = df_algs.loc[df_algs.index.isin(inds)]
            df.drop(expected_metrics+metrics+['AgeNBR'],axis=1).mean().sort_values(ascending=False)[:25].plot(kind='Bar')
            plt.show()


        r.data_source.on_change('selected', update)

HOWEVER

        if metric == 'LengthOfStayDaysNBR':
            color_mapper = linear_cmap(metric,palette=Spectral6,low=df_algs[metric].min(),high=df_algs[metric].max())
        else:
            color_mapper = factor_cmap(metric,palette=Spectral6,factors=df_algs[metric].unique())

seems to break the code. I get "Glyph refers to nonexistent column name: {}" when trying to color code the scatterplot.

Answer 1

I would not rely on a CustomJS callback that calls kernel.execute to do this. I would regard that as a very fragile way of doing things, that is susceptible to being confusing or unreasonable when cells are re-executed in an arbitrary order. Additionally, for future consideration, I am not sure it will ever work at all with the next generation JupyterLab, since kernel.execute may not be available.

Rather, I would embed a real Bokeh Server Application. The Bokeh server was expressly created to efficiently and robustly synchronize data between the JavaScript and Python parts of Bokeh over a defined and maintained protocol, and to provide for executing callbacks based on changes from either side.

Here is a screencast gif of an app embedded in a notebook that updates a pair of histograms based on the selection in the center plot. The code cells are given below the gif.

---

Cell 1

import numpy as np
from bokeh.io import output_notebook, show
from bokeh.layouts import row, column
from bokeh.models import BoxSelectTool, LassoSelectTool, Spacer
from bokeh.plotting import figure, curdoc

output_notebook()

Cell 2

x1 = np.random.normal(loc=5.0, size=400) * 100
y1 = np.random.normal(loc=10.0, size=400) * 10

x2 = np.random.normal(loc=5.0, size=800) * 50
y2 = np.random.normal(loc=5.0, size=800) * 10

x3 = np.random.normal(loc=55.0, size=200) * 10
y3 = np.random.normal(loc=4.0, size=200) * 10

x = np.concatenate((x1, x2, x3))
y = np.concatenate((y1, y2, y3))

TOOLS="pan,wheel_zoom,box_select,lasso_select,reset"

Cell 3

def app(doc):
    # create the scatter plot
    p = figure(tools=TOOLS, plot_width=600, plot_height=600, min_border=10, min_border_left=50,
               toolbar_location="above", x_axis_location=None, y_axis_location=None,
               title="Linked Histograms")
    p.background_fill_color = "#fafafa"
    p.select(BoxSelectTool).select_every_mousemove = False
    p.select(LassoSelectTool).select_every_mousemove = False

    r = p.scatter(x, y, size=3, color="#3A5785", alpha=0.6)

    # create the horizontal histogram
    hhist, hedges = np.histogram(x, bins=20)
    hzeros = np.zeros(len(hedges)-1)
    hmax = max(hhist)*1.1

    LINE_ARGS = dict(color="#3A5785", line_color=None)

    ph = figure(toolbar_location=None, plot_width=p.plot_width, plot_height=200, x_range=p.x_range,
                y_range=(-hmax, hmax), min_border=10, min_border_left=50, y_axis_location="right")
    ph.xgrid.grid_line_color = None
    ph.yaxis.major_label_orientation = np.pi/4
    ph.background_fill_color = "#fafafa"

    ph.quad(bottom=0, left=hedges[:-1], right=hedges[1:], top=hhist, color="white", line_color="#3A5785")
    hh1 = ph.quad(bottom=0, left=hedges[:-1], right=hedges[1:], top=hzeros, alpha=0.5, **LINE_ARGS)
    hh2 = ph.quad(bottom=0, left=hedges[:-1], right=hedges[1:], top=hzeros, alpha=0.1, **LINE_ARGS)

    # create the vertical histogram
    vhist, vedges = np.histogram(y, bins=20)
    vzeros = np.zeros(len(vedges)-1)
    vmax = max(vhist)*1.1

    pv = figure(toolbar_location=None, plot_width=200, plot_height=p.plot_height, x_range=(-vmax, vmax),
                y_range=p.y_range, min_border=10, y_axis_location="right")
    pv.ygrid.grid_line_color = None
    pv.xaxis.major_label_orientation = np.pi/4
    pv.background_fill_color = "#fafafa"

    pv.quad(left=0, bottom=vedges[:-1], top=vedges[1:], right=vhist, color="white", line_color="#3A5785")
    vh1 = pv.quad(left=0, bottom=vedges[:-1], top=vedges[1:], right=vzeros, alpha=0.5, **LINE_ARGS)
    vh2 = pv.quad(left=0, bottom=vedges[:-1], top=vedges[1:], right=vzeros, alpha=0.1, **LINE_ARGS)

    layout = column(row(p, pv), row(ph, Spacer(width=200, height=200)))

    doc.add_root(layout)
    doc.title = "Selection Histogram"

    def update(attr, old, new):
        inds = np.array(new['1d']['indices'])
        if len(inds) == 0 or len(inds) == len(x):
            hhist1, hhist2 = hzeros, hzeros
            vhist1, vhist2 = vzeros, vzeros
        else:
            neg_inds = np.ones_like(x, dtype=np.bool)
            neg_inds[inds] = False
            hhist1, _ = np.histogram(x[inds], bins=hedges)
            vhist1, _ = np.histogram(y[inds], bins=vedges)
            hhist2, _ = np.histogram(x[neg_inds], bins=hedges)
            vhist2, _ = np.histogram(y[neg_inds], bins=vedges)

        hh1.data_source.data["top"]   =  hhist1
        hh2.data_source.data["top"]   = -hhist2
        vh1.data_source.data["right"] =  vhist1
        vh2.data_source.data["right"] = -vhist2

    r.data_source.on_change('selected', update)

Cell 4

# set notebook_url appropriately
show(app, notebook_url="http://localhost:8889")