CODEWITHSUNDEEP
pythonpandasmatplotlibseaborn
Manuel
Manuel

Reputation: 802

Making several pieplots from dataframe columns

I made a dataframe from some data that looks like the following

df = pd.Dataframe({'Keys': ['key1','key2','key3']
                   'Percentage':[63,37,89]}

I want to create 3 piecharts that show the percentage for each key. Thought it was going to be easy but all I could find was how to create a chart that contains all 3 keys.

The quick way to fix it would be to create some lists for every key and make the chart based on that but is it there a way to do this more efficiently?

Edit:

I'd like to create something like this for each key

import matplotlib.pyplot as plt

fig1, ax1 = plt.subplots()
labels = ['key1_yes','key1_no']
sizes = [63,37]

ax1.set_title('key1')

ax1.pie(sizes, labels=labels, autopct='%1.1f%%',
        shadow=True, startangle=90)

Upvotes: 0

Views: 263

Answers (2)

SchwarzeHuhn
SchwarzeHuhn

Reputation: 648

If you define your df as follows:

import pandas as pd

df = pd.DataFrame({'key': ['key1', 'key1', 'key2', 'key2', 'key3', 'key3'],
                   'kind': ['yes', 'no', 'yes', 'no', 'yes', 'no'],
                   'n': [63, 37, 89, 11, 37, 63]})
df
    key kind   n
0  key1  yes  63
1  key1   no  37
2  key2  yes  89
3  key2   no  11
4  key3  yes  37
5  key3   no  63

then you can plot like this:

import matplotlib.pyplot as plt
import seaborn as sns

def my_pie(n, kind, **kwargs):
    plt.pie(x = n, labels = kind)

g = sns.FacetGrid(df, col = 'key')
g.map(my_pie, 'n', 'kind')

which yields

enter image description here

Upvotes: 2

Manuel
Manuel

Reputation: 802

Ended up doing something like this

import matplotlib.pyplot as plt

for index, row in new_data_df.iterrows():

    fig1, ax1 = plt.subplots(figsize=(5,5))
    labels = ['Yes','No']
    sizes = [row['Percentage'],row['1-Percentage']]

    ax1.set_title(row['Key'])

    ax1.pie(sizes, labels=labels, autopct='%1.2f%%', startangle=90, colors = ['gainsboro','dimgrey'])

Upvotes: 0

Related Questions