plotting a pandas dataframe row by row

Question

I have the following dataframe:

I want to create pie charts one for each row, the thing is that i am having trouble with the charts order, i want each chart to have a figsize of lets say 5,5 and that every row in my dataframe will be a row of plot in my subplots with the index as title.

tried many combinations and playing with pyploy.subplots but not success. would be glad for some help.

Thanks

Answer 1

You can either transpose your dataframe and using pandas pie kind for plotting, i.e. df.transpose().plot(kind='pie', subplots=True) or iterate through rows while sub plotting.

An example using subplots:

import pandas as pd
import matplotlib.pyplot as plt
import numpy as np

# Recreate a similar dataframe
rows = ['rows {}'.format(i) for i in range(5)]
columns = ['hits', 'misses']
col1 = np.random.random(5)
col2 = 1 - col1
data = zip(col1, col2)

df = pd.DataFrame(data=data, index=rows, columns=columns)

# Plotting

fig = plt.figure(figsize=(15,10))

for i, (name, row) in enumerate(df.iterrows()):
    ax = plt.subplot(2,3, i+1)
    ax.set_title(row.name)
    ax.set_aspect('equal')
    ax.pie(row, labels=row.index)

plt.show()