How to annotate each segment of a stacked bar chart

Imports and DataFrame

import pandas as pd
import matplotlib.pyplot as plt

data = {'var': ['TR', 'AC', 'F&B'], '2019 1Q': [6600, 1256, 588], '2019 2Q': [6566, 1309, 586], '2019 3Q': [7383, 1525, 673]}
df = pd.DataFrame(data)
df.set_index('var', inplace=True)

# display(df)
     2019 1Q  2019 2Q  2019 3Q
var                           
TR      6600     6566     7383
AC      1256     1309     1525
F&B      588      586      673

Update as of matplotlib v3.4.2

• Use matplotlib.pyplot.bar_label
• See How to add value labels on a bar chart for additional details and examples with .bar_label.
• Tested with pandas v1.2.4, which is using matplotlib as the plot engine.

ax = df.T.plot.bar(stacked=True, figsize=(6, 5), rot=0)

for c in ax.containers:
    ax.bar_label(c, label_type="center")
    
ax.legend(title="Categories", bbox_to_anchor=(1.05, 1), loc="upper left")

Original Answer – prior to matplotlib v3.4.2

• Transpose the dataframe and then use pandas.DataFrame.plot.bar with stacked=True.
• An ndarray is returned with one matplotlib.axes.Axes per column with subplots=True.
  • In the case of this figure, ax.patches contains 9 matplotlib.patches.Rectangle objects, one for each segment of each bar.
    • By using the associated methods for this object, the height, width, x, and y locations can be extracted, and used to annotate the rectangles.
• The difference this question has from How to annotate a stacked bar chart with word count and column name? is the other question needs to extract and use alternate text for labels, and this dataframe needed to be transposed.

ax = df.T.plot.bar(stacked=True)
plt.legend(title="Categories", bbox_to_anchor=(1.05, 1), loc="upper left")

for i, rect in enumerate(ax.patches):
    # Find where everything is located
    height = rect.get_height()
    width = rect.get_width()
    x = rect.get_x()
    y = rect.get_y()

    # The height of the bar is the count value and can used as the label
    label_text = f'{height:.0f}'

    label_x = x + width / 2
    label_y = y + height / 2

    # don't include label if it's equivalently 0
    if height > 0.001:
        ax.text(label_x, label_y, label_text, ha="center", va="center", fontsize=8)