Making multiple pie charts out of a pandas dataframe (one for each row)

Question

I have a dataframe (df) that shows emotions associated with various categories of business:

My task is to create pie charts showing the % of emotions for each type of business. So I need to create a function in matplotlib that reads the "Business" column and then builds a pie chart using each of the emotion categories for each row in the dataframe.

I've already built a bar plot, but I am having no luck with the pie chart. EDIT: HERE IS MY CODE FOR THE BAR PLOT:

import pandas as pd
import csv
import matplotlib.pyplot as plt
GraphData = open("barGraph.csv")
df = pd.read_csv('barGraph.csv')
ax = df.plot(kind='bar', title ="Emotions at Various Businesses", figsize=(15, 10), legend=True, fontsize=12)
ax.set_xlabel("Business Type",fontsize=12)
ax.set_ylabel("Strength of Emotion",fontsize=12)
ax.set_xticklabels(['Beauty & Spas', 'Burgers-Restaurants', 'Pizza', 'Mexican Restaurants', 'Modern European-Restaurants', 'Chinese'])
plt.show()

I've read the documentation on pie charts, but it isn't making sense to me, at least as it pertains to drawing the data from a dataframe as opposed to a series.

Any suggestions?

Maybe this is helpful: [Plotting Pandas DataFrames in to Pie Charts using matplotlib](http://stackoverflow.com/questions/21090316/plotting-pandas-dataframes-in-to-pie-charts-using-matplotlib) — Lucas, Jan 22 '17 at 04:35

score 11 · Accepted Answer · answered Jan 22 '17 at 06:27

11

Consider the dataframe df

df = pd.DataFrame(dict(
        Business='Beauty & Spas;Burgers-Restaurants;Pizza;Mexican Restaurants;Modern European-Restaurants;Chineese'.split(';'),
        aniticipation=[0] * 6,
        enjoyment=[6., 1., 6., 33.,150., 19.5],
        sad=[1., 2., 1., 3., 13.5, 0.],
        disgust=[1, 1, 0, 3, 37, 3],
        anger=[1.5, 2., 4., 9., 19., 3.],
        surprise=[3, 0, 0, 2, 12, 1],
        fear=[0, 1, 1, 9, 22, 1],
        trust=[0] * 6
    ))

You can create pie charts like this

fig, axes = plt.subplots(2, 3, figsize=(10, 6))

for i, (idx, row) in enumerate(df.set_index('Business').iterrows()):
    ax = axes[i // 3, i % 3]
    row = row[row.gt(row.sum() * .01)]
    ax.pie(row, labels=row.index, startangle=30)
    ax.set_title(idx)

fig.subplots_adjust(wspace=.2)

answered Jan 22 '17 at 06:27

piRSquared

285,575
57
475
624

Thanks, piRSquared. I really appreciate it. – Andrew Smith Jan 22 '17 at 14:17
@AndrewSmith you're probably using jupyter notebook and you need to run '%matplotlib inline' – piRSquared Jan 22 '17 at 14:18
Yes, exactly. Why does Jupyter require the inline format? – Andrew Smith Jan 22 '17 at 14:19
@piRSquared , It was nice code, thanks. if I want to display values too in the each pie. how can we do that ? For ex : Beauty & Spas, we want to display enjoyment is 6 which is red color. Any help? – Jun 28 '17 at 04:49
Great Work, worked for me, I was wondering how you give a title for the whole pie chart? – Shahriar Kabir Khan Feb 23 '21 at 01:41

Making multiple pie charts out of a pandas dataframe (one for each row)

1 Answers1

Linked