Matplotlib Histograms

In Matplotlib, histograms are used to plot the distribution of a dataset by dividing the data into bins and counting the number of observations in each bin. This type of plot is useful for visualizing the frequency distribution of continuous or discrete data.

You can create histograms using the plt.hist() function. This function automatically calculates the bins and plots the frequency of values within each bin.

1. Basic Histogram

A basic histogram can be created by passing a dataset to plt.hist().

Example: Basic Histogram

import matplotlib.pyplot as plt
import numpy as np

# Generate random data
data = np.random.randn(1000)

# Create a basic histogram
plt.hist(data)

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.title('Basic Histogram')

# Show the plot
plt.show()

In this example:

• The dataset data is generated from a normal distribution.
• plt.hist() automatically creates the bins and counts the number of observations in each bin.

2. Specifying the Number of Bins

By default, Matplotlib decides the number of bins, but you can manually specify the number of bins using the bins parameter.

Example: Histogram with Custom Number of Bins

# Create a histogram with custom number of bins
plt.hist(data, bins=20)

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.title('Histogram with 20 Bins')

# Show the plot
plt.show()

Here, the bins=20 argument creates 20 equally spaced bins.

3. Customizing Bin Edges

You can define custom bin edges by passing a list or array to the bins parameter.

Example: Histogram with Custom Bin Edges

# Define custom bin edges
bin_edges = [-3, -2, -1, 0, 1, 2, 3]

# Create a histogram with custom bin edges
plt.hist(data, bins=bin_edges)

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.title('Histogram with Custom Bin Edges')

# Show the plot
plt.show()

In this example, the bins are defined by the bin_edges list, with edges at -3, -2, -1, 0, 1, 2, and 3.

4. Changing Histogram Colors

You can customize the color of the bars using the color parameter, and the edge color of the bars using the edgecolor parameter.

Example: Histogram with Custom Colors

# Create a histogram with custom colors
plt.hist(data, bins=20, color='skyblue', edgecolor='black')

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.title('Histogram with Custom Colors')

# Show the plot
plt.show()

This creates a histogram where the bars are filled with the color 'skyblue', and the edges of the bars are outlined in 'black'.

5. Normalizing the Histogram

By default, histograms show the raw frequency counts. To display the probability density (i.e., normalize the histogram so that the area under the bars sums to 1), you can set density=True.

Example: Normalized Histogram

# Create a normalized histogram
plt.hist(data, bins=20, density=True, color='lightgreen')

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Density')
plt.title('Normalized Histogram')

# Show the plot
plt.show()

Here, the density=True argument ensures that the total area under the histogram is 1, and the y-axis represents the probability density instead of frequency.

6. Adding a Histogram Outline (Step Plot)

You can create a histogram outline or step plot by setting the histtype='step' parameter. This removes the filled bars and only shows the outlines.

Example: Step Histogram

# Create a step histogram
plt.hist(data, bins=20, histtype='step', color='red')

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.title('Step Histogram')

# Show the plot
plt.show()

This creates a histogram where only the outlines of the bins are drawn.

7. Cumulative Histogram

A cumulative histogram shows the cumulative frequency or cumulative probability. You can create a cumulative histogram by setting cumulative=True.

Example: Cumulative Histogram

# Create a cumulative histogram
plt.hist(data, bins=20, cumulative=True, color='purple')

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Cumulative Frequency')
plt.title('Cumulative Histogram')

# Show the plot
plt.show()

In this example, the cumulative=True argument causes the histogram to display cumulative frequencies, where each bar represents the total frequency up to that point.

8. Multiple Histograms on the Same Plot

You can plot multiple histograms on the same plot by calling plt.hist() multiple times with different datasets and adjusting the alpha parameter to control the transparency.

Example: Multiple Histograms

# Generate two datasets
data1 = np.random.randn(1000)
data2 = np.random.randn(1000) + 2  # Shifted by 2

# Create overlapping histograms
plt.hist(data1, bins=20, alpha=0.5, label='Dataset 1', color='blue')
plt.hist(data2, bins=20, alpha=0.5, label='Dataset 2', color='green')

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency')
plt.title('Overlapping Histograms')

# Add legend
plt.legend()

# Show the plot
plt.show()

In this example:

• Two datasets (data1 and data2) are plotted on the same axes.
• The alpha=0.5 argument makes the bars semi-transparent to allow for overlap.
• The label parameter is used to identify each dataset, and plt.legend() adds a legend.

9. 2D Histogram (Hexbin Plot)

For two-dimensional data, you can create a 2D histogram or hexbin plot using the plt.hexbin() function. This groups data points into hexagonal bins and counts the number of observations in each bin.

Example: 2D Histogram

# Generate 2D data
x = np.random.randn(1000)
y = np.random.randn(1000)

# Create a 2D histogram (hexbin plot)
plt.hexbin(x, y, gridsize=30, cmap='Blues')

# Add colorbar
plt.colorbar(label='Count')

# Add labels and title
plt.xlabel('X')
plt.ylabel('Y')
plt.title('2D Histogram (Hexbin Plot)')

# Show the plot
plt.show()

In this example:

• plt.hexbin() creates a 2D histogram, where the data is grouped into hexagonal bins.
• gridsize controls the number of hexagons, and cmap specifies the color map.
• plt.colorbar() adds a color bar to indicate the count in each bin.

10. Histogram with Logarithmic Scale

You can apply a logarithmic scale to the y-axis of a histogram using plt.yscale().

Example: Histogram with Logarithmic Scale

# Create a histogram with a logarithmic y-axis
plt.hist(data, bins=20, color='orange')

# Apply logarithmic scale to the y-axis
plt.yscale('log')

# Add labels and title
plt.xlabel('Value')
plt.ylabel('Frequency (log scale)')
plt.title('Histogram with Logarithmic Scale')

# Show the plot
plt.show()

In this example, the y-axis is displayed using a logarithmic scale, which is useful for visualizing data with large variations in frequency.

Conclusion

Histograms are a powerful tool for visualizing the distribution of data in Matplotlib. With various customizations such as bin sizes, colors, normalization, and cumulative options, you can create histograms that fit your data analysis needs. Additionally, advanced options like step histograms, 2D histograms, and multiple histograms on the same plot allow for deeper exploration of datasets.