How do I create a pie chart using categorical data in matplotlib?

user_12749240 0
Viewed 16082

I have data as follows:

ID  Gender  Country  ...
1   Male    UK
2   Female  US
3   Male    NZ
4   Female  UK
...

There are only 2 options for gender and 3 for country. I would like to create a seperate pie chart for both "Gender" and "Country" to show how many times each option shows up in the data but I'm quite confused about how to do so.

The data is stored in a pandas dataframe.

Any and all help is much appreciated!

pie-chart pandas matplotlib python
3 Answers

Here is an approach using pandas:

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

def label_function(val):
    return f'{val / 100 * len(df):.0f}\n{val:.0f}%'

N = 50
df = pd.DataFrame({'country': np.random.choice(['UK', 'US', 'NZ'], N),
                   'gender': np.random.choice(['Male', 'Female'], N)})

fig, (ax1, ax2) = plt.subplots(ncols=2, figsize=(10, 5))

df.groupby('country').size().plot(kind='pie', autopct=label_function, textprops={'fontsize': 20},
                                  colors=['tomato', 'gold', 'skyblue'], ax=ax1)
df.groupby('gender').size().plot(kind='pie', autopct=label_function, textprops={'fontsize': 20},
                                 colors=['violet', 'lime'], ax=ax2)
ax1.set_ylabel('Per country', size=22)
ax2.set_ylabel('Per gender', size=22)plt.tight_layout()
plt.show()

example plot

PS: To just show the percentage, use autopct='%1.0f%%'.

user_120464090

I assume you started with this

import pandas as pd
from matplotlib.pyplot import pie, axis, show

df = pd.DataFrame([[1,'Male','UK'],   [2, 'Female', 'NZ'],    [3, 'Male', 'UK'], [4, 'Male', 'US']], columns=['ID',  'Gender',  'Country'])

Plot for gender

df.groupby('gender').size().plot(kind='pie', autopct='%.2f')

enter image description here

Plot for country

df.groupby('country').size().plot(kind='pie', autopct='%.2f')

enter image description here

user_113927600

Okay so since you are using a dataframe like this:

data = pd.DataFrame([[1,'Male','UK'],   [2, 'Female', 'NZ'],    [3, 'Male', 'UK'], [4, 'Male', 'US']], columns=['ID',  'Gender',  'Country'])

You can really just do:

data['Gender'].value_counts().plot(kind='pie')

If you wanna do it manually:

people = len(data.Gender)
genders = len(set(data.Gender))

res = []
for gender in set(data.Gender):
    res.append([gender, len(data[data['Gender']==gender]), len(data[data['Gender']==gender])/people])

and then just plot it.

user_80848050
Related