How do I plot a histogram with grouped data in matplotlib?

General Discussions
,
1

I know how to plot a histogram when individual datapoints are available, like:

matplotlib.pyplot.hist(x, bins=10)

But what if I only have grouped data, such as:

Marks Number of students
0-10 8
10-20 12
20-30 24
30-40 26
… …

I understand that I can use a bar plot to mimic a histogram by adjusting the x-ticks, but is it possible to create this using the hist function of matplotlib.pyplot?

2

I’ve worked with this kind of data representation a few times. If you want to plot a histogram in histogram Python matplotlib with grouped data, you can use the hist function with predefined bin edges. It’s pretty straightforward once you get the hang of it.

Here’s how you can do it:

import matplotlib.pyplot as plt

# Grouped data: intervals and counts
bins = [0, 10, 20, 30, 40]
counts = [8, 12, 24, 26]

# Plotting the histogram
plt.hist([10] * counts[0] + [15] * counts[1] + [25] * counts[2] + [35] * counts[3], bins=bins, edgecolor='black')
plt.show()

This method essentially creates synthetic data that fits your grouped data so it can be used directly by hist to plot the histogram. It’s a neat trick when you’re working with grouped intervals!

3

That’s a great way, Dimple! Another approach, which I’ve often found useful when working with histogram Python matplotlib, is to use a bar plot with custom x-ticks and widths. It keeps things closer to the grouped data without having to create synthetic points.

Here’s an example:

import matplotlib.pyplot as plt
import numpy as np

# Grouped data: intervals and counts
labels = ['0-10', '10-20', '20-30', '30-40']
counts = [8, 12, 24, 26]

# Define the positions and width for the bars
x = np.arange(len(labels))
width = 0.8

# Bar plot
plt.bar(x, counts, width, edgecolor='black')

# Customize x-ticks
plt.xticks(x, labels)

plt.show()

This method uses a bar plot to represent the data while preserving the histogram feel by aligning the x-ticks with the intervals. It’s especially helpful when you want to emphasize the grouped nature of your data.

4

Good call, Fathima! That works well for a grouped feel. If you’re like me and prefer sticking to the histogram approach while avoiding synthetic data, you can use the weights argument in histogram Python matplotlib. It directly represents the frequency of each bin.

Here’s how:

import matplotlib.pyplot as plt
import numpy as np

# Grouped data: intervals and counts
bins = [0, 10, 20, 30, 40]
counts = [8, 12, 24, 26]

# Data to fit the bins
data = np.array([0] * 8 + [10] * 12 + [20] * 24 + [30] * 26)

# Plotting the histogram using weights
plt.hist(data, bins=bins, edgecolor='black')
plt.show()

With this method, you use weights to directly reflect the frequencies for each bin. It’s perfect when you want your histogram to match the grouped data without synthesizing too much.