Ctrl+K

Bar Charts

Bar Charts#

Let’s say we want to create a bar chart with a single line of data from our df DataFrame. We can select a specific row using iloc.

import pandas as pd, matplotlib.pyplot as plt, numpy as np # import statement
df = pd.read_csv('https://raw.githubusercontent.com/kwaldenphd/elements-of-computing/main/book/data/ch10/air_quality_no2.csv', index_col=0, parse_dates=True) # load data

# create bar plot for single row of data
df.iloc[5].plot.bar()

<Axes: >
../_images/0d162528e784a0efebb66b7c8aab56959dd6308a7f49f2a14dc736927332bf7b.png

Grouped Bar Chart#

Let’s say we want to produce a grouped bar chart. We can create a DataFrame with four columns and random values. By default, .plot.bar() creates a separate bar for each of the numeric columns. We could select a single column using [''] to only show one bar or column value.

df2 = pd.DataFrame(np.random.randn(10, 4), columns=['a', 'b', 'c', 'd']) # create dataframe
df2.plot.bar() # grouped bar chart with categorical data

<Axes: >
../_images/d3224443ac8c0332fab6213874ca9469b603e3a479e062f53b38aa396cb89027.png

Stacked Bar Chart#

If we wanted to show this data as a stacked bar chart, we would set the stacked parameter to True.

df2.plot.bar(stacked=True) # stacked bar chart

Horizontal Bar Chart#

Remember .plot.barh() generates a horizontal bar chart, and we can also set stacked to True here to create a horizontal stacked bar chart.

df2.plot.barh(stacked=True) # horizontal bar chart

Additional Resources#

For more on bar charts:

Contents