Create a Hat Graph

Introduction

In this lab, we will learn how to create a Hat Graph in Python using the Matplotlib library. A Hat Graph is a variant of a stacked bar chart, where each bar is a hat shape. We will use a dataset of scores by number of game and players to create the graph.

VM Tips

After the VM startup is done, click the top left corner to switch to the Notebook tab to access Jupyter Notebook for practice.

Sometimes, you may need to wait a few seconds for Jupyter Notebook to finish loading. The validation of operations cannot be automated because of limitations in Jupyter Notebook.

If you face issues during learning, feel free to ask Labby. Provide feedback after the session, and we will promptly resolve the problem for you.

Skills Graph

Import Libraries

In this step, we will import the necessary libraries for creating the Hat Graph.

import matplotlib.pyplot as plt
import numpy as np

Define the Hat Graph Function

In this step, we will define a function that creates the Hat Graph. The function takes in the following parameters:

• ax: The Axes to plot into.
• xlabels: The category names to be displayed on the x-axis.
• values: The data values. Rows are the groups, and columns are the categories.
• group_labels: The group labels displayed in the legend.

def hat_graph(ax, xlabels, values, group_labels):
    """
    Create a hat graph.

    Parameters
    ----------
    ax : matplotlib.axes.Axes
        The Axes to plot into.
    xlabels : list of str
        The category names to be displayed on the x-axis.
    values : (M, N) array-like
        The data values.
        Rows are the groups (len(group_labels) == M).
        Columns are the categories (len(xlabels) == N).
    group_labels : list of str
        The group labels displayed in the legend.
    """

    def label_bars(heights, rects):
        """Attach a text label on top of each bar."""
        for height, rect in zip(heights, rects):
            ax.annotate(f'{height}',
                        xy=(rect.get_x() + rect.get_width() / 2, height),
                        xytext=(0, 4),  ## 4 points vertical offset.
                        textcoords='offset points',
                        ha='center', va='bottom')

    values = np.asarray(values)
    x = np.arange(values.shape[1])
    ax.set_xticks(x, labels=xlabels)
    spacing = 0.3  ## spacing between hat groups
    width = (1 - spacing) / values.shape[0]
    heights0 = values[0]
    for i, (heights, group_label) in enumerate(zip(values, group_labels)):
        style = {'fill': False} if i == 0 else {'edgecolor': 'black'}
        rects = ax.bar(x - spacing/2 + i * width, heights - heights0,
                       width, bottom=heights0, label=group_label, **style)
        label_bars(heights, rects)

Prepare Data

In this step, we will initialise labels and a numpy array. We will make sure we have N labels of N number of values in the array.

## initialise labels and a numpy array make sure you have
## N labels of N number of values in the array
xlabels = ['I', 'II', 'III', 'IV', 'V']
playerA = np.array([5, 15, 22, 20, 25])
playerB = np.array([25, 32, 34, 30, 27])

Create Hat Graph

In this step, we will create the Hat Graph using the data prepared in the previous step and the hat_graph function.

fig, ax = plt.subplots()
hat_graph(ax, xlabels, [playerA, playerB], ['Player A', 'Player B'])

## Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_xlabel('Games')
ax.set_ylabel('Score')
ax.set_ylim(0, 60)
ax.set_title('Scores by number of game and players')
ax.legend()

fig.tight_layout()
plt.show()

Complete Code

Here is the complete code to create the Hat Graph in Python.

import matplotlib.pyplot as plt
import numpy as np


def hat_graph(ax, xlabels, values, group_labels):
    """
    Create a hat graph.

    Parameters
    ----------
    ax : matplotlib.axes.Axes
        The Axes to plot into.
    xlabels : list of str
        The category names to be displayed on the x-axis.
    values : (M, N) array-like
        The data values.
        Rows are the groups (len(group_labels) == M).
        Columns are the categories (len(xlabels) == N).
    group_labels : list of str
        The group labels displayed in the legend.
    """

    def label_bars(heights, rects):
        """Attach a text label on top of each bar."""
        for height, rect in zip(heights, rects):
            ax.annotate(f'{height}',
                        xy=(rect.get_x() + rect.get_width() / 2, height),
                        xytext=(0, 4),  ## 4 points vertical offset.
                        textcoords='offset points',
                        ha='center', va='bottom')

    values = np.asarray(values)
    x = np.arange(values.shape[1])
    ax.set_xticks(x, labels=xlabels)
    spacing = 0.3  ## spacing between hat groups
    width = (1 - spacing) / values.shape[0]
    heights0 = values[0]
    for i, (heights, group_label) in enumerate(zip(values, group_labels)):
        style = {'fill': False} if i == 0 else {'edgecolor': 'black'}
        rects = ax.bar(x - spacing/2 + i * width, heights - heights0,
                       width, bottom=heights0, label=group_label, **style)
        label_bars(heights, rects)


## initialise labels and a numpy array make sure you have
## N labels of N number of values in the array
xlabels = ['I', 'II', 'III', 'IV', 'V']
playerA = np.array([5, 15, 22, 20, 25])
playerB = np.array([25, 32, 34, 30, 27])

fig, ax = plt.subplots()
hat_graph(ax, xlabels, [playerA, playerB], ['Player A', 'Player B'])

## Add some text for labels, title and custom x-axis tick labels, etc.
ax.set_xlabel('Games')
ax.set_ylabel('Score')
ax.set_ylim(0, 60)
ax.set_title('Scores by number of game and players')
ax.legend()

fig.tight_layout()
plt.show()

Summary

In this lab, we learned how to create a Hat Graph in Python using the Matplotlib library. We defined a function that creates the Hat Graph, prepared the data, and created the graph using the hat_graph function. The Hat Graph is a variant of a stacked bar chart, where each bar is a hat shape. We used a dataset of scores by number of game and players to create the graph.