Creating Pandas DataFrame from the data points selected on the Plotly Scatter Plot

Question

Plotly Figure Widget helps me create a scatter plot which is interactive, ie., I can select data points on the scatter plot and based on the selection my table widget shows the records. I wanted help with converting this table to a pandas dataframe.

import plotly.graph_objs as go
import plotly.offline as py

import pandas as pd
import numpy as np
from ipywidgets import interactive, HBox, VBox

py.init_notebook_mode()

df = pd.read_csv('https://raw.githubusercontent.com/jonmmease/plotly_ipywidget_notebooks/master/notebooks/data/cars/cars.csv')

f = go.FigureWidget([go.Scatter(y = df['City mpg'], x = df['City mpg'], mode = 'markers')])
scatter = f.data[0]
N = len(df)
scatter.x = scatter.x + np.random.rand(N)/10 *(df['City mpg'].max() - df['City mpg'].min())
scatter.y = scatter.y + np.random.rand(N)/10 *(df['City mpg'].max() - df['City mpg'].min())
scatter.marker.opacity = 0.5

def update_axes(xaxis, yaxis):
    scatter = f.data[0]
    scatter.x = df[xaxis]
    scatter.y = df[yaxis]
    with f.batch_update():
        f.layout.xaxis.title = xaxis
        f.layout.yaxis.title = yaxis
        scatter.x = scatter.x + np.random.rand(N)/10 *(df[xaxis].max() - df[xaxis].min())
        scatter.y = scatter.y + np.random.rand(N)/10 *(df[yaxis].max() - df[yaxis].min())

axis_dropdowns = interactive(update_axes, yaxis = df.select_dtypes('int64').columns, xaxis = df.select_dtypes('int64').columns)

# Create a table FigureWidget that updates on selection from points in the scatter plot of f
t = go.FigureWidget([go.Table(
    header=dict(values=['ID','Classification','Driveline','Hybrid'],
                fill = dict(color='#C2D4FF'),
                align = ['left'] * 5),
    cells=dict(values=[df[col] for col in ['ID','Classification','Driveline','Hybrid']],
               fill = dict(color='#F5F8FF'),
               align = ['left'] * 5))])

def selection_fn(trace,points,selector):
    t.data[0].cells.values = [df.loc[points.point_inds][col] for col in ['ID','Classification','Driveline','Hybrid']]

scatter.on_selection(selection_fn)

# Put everything together
VBox((HBox(axis_dropdowns.children),f,t))

Just expecting the table created after selecting points on the scatter plot to a pandas dataframe.

Output

Answer 1

Probably not the most elegant way to solve it, but after you select your points, you can type:

d = t.to_dict()
df = pd.DataFrame(d['data'][0]['cells']['values'], index =d['data'][0]['header']['values']).T

t is of type plotly.graph_objs._figurewidget.FigureWidget

I use jupyter notebook, so I wrote these lines of code one cell below your code, and I get a new df with the selected events

Answer 2

Assuming the following piece of code highlights the points you care about:

def selection_fn(trace,points,selector):
    t.data[0].cells.values = [df.loc[points.point_inds][col] for col in ['ID','Classification','Driveline','Hybrid']]

Change it to return a dataframe:

def selection_fn(trace,points,selector):
    return pd.df([df.loc[points.point_inds][col] for col in ['ID','Classification','Driveline','Hybrid'] if col in {selection}])

The list comprehension needs to be changed to loop over only the points you want to return. Example list comphrension from the documentation:

[(x, y) for x in [1,2,3] for y in [3,1,4] if x != y]

Answer 3

A better solution:

    def selection_fn(trace, points, selector):
    
        t.data[0].cells.values = [
            df.loc[points.point_inds][col]
            for col in ["ID", "Classification", "Driveline", "Hybrid"]]
        
        selection_fn.df1 = df.loc[points.point_inds]


    print(selection_fn.df1)

Access a function variable outside the function without using "global"