In a Plotly histogram, how to make each animation frame a row in the dataframe?

Question

I have the following pandas dataframe:

       month  stories  comments  comment_authors  story_authors
0    2006-10       49        12                4             16
1    2007-02      564       985              192            163
2    2007-03     1784      4521              445            287

I am trying to construct a Plotly histogram where there are four categorical bins (x-axis) for each of the stories, comments, comment_authors, and story_authors columns, and the count (y-axis) is the given quantity for a specific month (i.e. a specific row). I am then trying to animate the histogram based on month in Plotly Express using animation_frame and animation_group.

For example, the first histogram for month=2006-10 would look something like:

50 |  ____
45 | |    |
40 | |    |
35 | |    |
30 | |    |
25 | |    |
20 | |    |
15 | |    |                               ____
10 | |    |    ____                      |    |
 5 | |    |   |    |    ______           |    |
 0  ----------------------------------------------------
     stories  comments  comment_authors  story_authors

In the histogram for the next animation frame, it would read the values from the stories, comments, comment_authors, story_authors columns for month=2007-02.

Is this possible to construct in Plotly? Is there a better figure to use than px.Histogram, like px.Bar? I have tried putting the columns on the x-axis and using the month for the animation frame, but this stacks the columns into one bin on the histogram and uses the count of the entire column, not a specific row's value.

histogram = dcc.Graph(figure=px.histogram(
    df, x=['stories', 'comments', 'comment_authors', 'story_authors'],
    animation_frame='month', animation_group='month'
))

Answer 1

It is not possible to draw a histogram with the data you are presenting. The best you can do is a bar chart, and you can animate it with a time series. I have modified the sample in the official reference to fit your data.

import pandas as pd
import numpy as np
import io

data = '''
      month  stories  comments  comment_authors  story_authors
0    2006-10       49        12                4             16
1    2007-02      564       985              192            163
2    2007-03     1784      4521              445            287
'''

df = pd.read_csv(io.StringIO(data), delim_whitespace=True)
df.set_index('month', inplace=True)
dfs = df.unstack().reset_index()
dfs.columns = ['category', 'month', 'value']
import plotly.express as px

fig = px.bar(dfs, x='category', y='value', color='category', animation_frame='month', range_y=[0,dfs['value'].max()])

fig.show()