How to plot subplots from a condition applied on a single column and the data available on another s-CodePudding

I have a dataframe which is much like the one following:

data = {'A':[21,22,23,24,25,26,27,28,29,30,11,12,13,14,15,16,17,18,19,20,1,2,3,4,5,6,7,8,9,10],
        'B':[8,8,8,8,8,8,8,8,8,8,5,5,5,5,5,5,5,5,5,5,3,3,3,3,3,3,3,3,3,3],
        'C':[10,15,23,17,18,26,24,30,35,42,44,42,38,36,34,30,27,25,27,24,1,0,2,3,5,26,30,40,42,50]}

data_df = pd.DataFrame(data)
data_df

I would like to have the subplots, the number of subplots should be equal to number of unique values of column 'B'. X axis = Values in column 'A' and Y axis = values in Column 'C'.

The code that I tried:

fig = px.line(data_df, 
              x='A', 
              y='C', 
              color='B',
              facet_col = 'B',
              )
fig.show()

gives output like

However, I would like to have the graphs in a single column, each graph autoscaled to the relevant area and resolution on the axes. Possibility: Can I somehow make use of groupby command to do it?

Since I may have other number of unique values in column 'B' (for example 5 unique values) based on other data, I would like to have this piece of code to work dynamic. Kindly help me.

PS: plotly express module is used to plot the graph.

CodePudding user response：

In order to stack all subplot in one column, and make sure that each xaxis is independent, just add the following in your px.line() call:

facet_col_wrap=1

And then follow up with:

fig.update_xaxes(matches=None)

Plot 1

Complete code:

import plotly.express as px
import pandas as pd
data = {'A':[21,22,23,24,25,26,27,28,29,30,11,12,13,14,15,16,17,18,19,20,1,2,3,4,5,6,7,8,9,10],
        'B':[8,8,8,8,8,8,8,8,8,8,5,5,5,5,5,5,5,5,5,5,3,3,3,3,3,3,3,3,3,3],
        'C':[10,15,23,17,18,26,24,30,35,42,44,42,38,36,34,30,27,25,27,24,1,0,2,3,5,26,30,40,42,50]}

data_df = pd.DataFrame(data)
data_df

fig = px.line(data_df, 
              x='A', 
              y='C', 
              color='B',
              facet_col = 'B',
              facet_col_wrap=1
             )
fig.update_xaxes(matches=None)
fig.show()

CodePudding user response：

You can instead use the argument facet_row = 'B' which will automatically stack the subplots by rows. Then to automatically rescale, you'll want to set all of the x data to the same array of values, which can be done by looping through fig.data and modifying fig.data[i]['x'] for each i.

import pandas as pd
import plotly.express as px

data = {'A':[21,22,23,24,25,26,27,28,29,30,11,12,13,14,15,16,17,18,19,20,1,2,3,4,5,6,7,8,9,10],
        'B':[8,8,8,8,8,8,8,8,8,8,5,5,5,5,5,5,5,5,5,5,3,3,3,3,3,3,3,3,3,3],
        'C':[10,15,23,17,18,26,24,30,35,42,44,42,38,36,34,30,27,25,27,24,1,0,2,3,5,26,30,40,42,50]}

data_df = pd.DataFrame(data)
fig = px.line(data_df, 
              x='A', 
              y='C', 
              color='B',
              facet_row = 'B',
              )
for fig_data in fig.data:
    fig_data['x'] = list(range(len(fig_data['y'])))
fig.show()