How do I add bar labels in each catplot facet with horizontal bars?

Question

I had Facet Grids and managed to make them horizontal, however after that I have searched for an appropriate code to display the percentages as before but what I found just isn't working as expected.

import seaborn as sns
import pandas as pd
import numpy as np

data = {
'id': [1, 1, 2, 2, 3, 3, 4, 4, 5, 5, 6, 6, 7, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 13, 13, 14, 14, 15, 15, 16, 16, 17, 17, 18, 18, 19, 19, 20, 20],
'survey': ['baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline', 'baseline', 'endline'],
'level': ['low', 'high', 'medium', 'low', 'high', 'medium', 'medium', 'high', 'low', 'low', 'medium', 'high', 'low', 'medium', 'low', 'high', 'low', 'low', 'medium', 'high', 'high', 'high', 'high', 'medium', 'low', 'low', 'medium', 'high', 'low', 'medium', 'high', 'medium', 'low', 'high', 'high', 'medium', 'medium', 'low', 'high', 'low']
}

df = pd.DataFrame(data)
df.survey.value_counts()

# Prep data
df = (
df.groupby(["survey", "level"])
.count()
.sort_index(ascending=False)
.rename(columns = {"id": "count"})
)
# Add percentages
pcts = []
for idx in ["endline", "baseline"]:
    pcts = (
        pcts + (df.loc[(idx,), "count"] * 100 / df.loc[(idx,), "count"].sum()).tolist()
    )

df["pct"] = pcts
df = df.reset_index()


df

g = sns.FacetGrid(df, col = "survey")
g.map_dataframe(sns.barplot,'pct', 'level', plt.margins(x = 0.5), hue = 'level', palette = 'Set1', orient = 'h')
g.set_axis_labels('Percentage', 'level') 


def annotate_bars(ax=None, fmt='.2f', **kwargs):
    ax = plt.gca() if ax is None else ax
    total = len(df['pct'])
    for p in ax.patches:
        percentage = '{:.1f}%'.format(100 * p.get_width()/total)
        x = p.get_x() + p.get_width() + 0.02
        y = p.get_y() + p.get_height()/2
        ax.annotate(percentage, (x, y))
g.map(annotate_bars, fmt='.2g', fontsize=8, color='k')
g.set_titles(col_template = '{col_name} survey')

score 1 · Answer 1 · answered Feb 04 '23 at 12:42

You're close, you just need to compute the sum/2 of % in df['pct'] rather than its length.

def annotate_bars(ax=None, fmt='.2f', **kwargs):
    ax = plt.gca() if ax is None else ax
    total = df['pct'].sum()/2            # <- this is the line you need to modify
    for p in ax.patches:
        percentage = '{:.1f}%'.format(100 * p.get_width()/total)
        x = p.get_x() + p.get_width() + 0.02
        y = p.get_y() + p.get_height()/2
        ax.annotate(percentage, (x, y))
g.map(annotate_bars, fmt='.2g', fontsize=8, color='k')
g.set_titles(col_template = '{col_name} survey')

Output :

Trenton McKinney · Answer 2 · 2023-02-04T19:18:20.000

As per the seaborn.FacetGrid documentation, use figure-level functions like sns.catplot.
The correct way to annotate bars is with matplotlib.pyplot.bar_label, which is thoroughly explained in this answer.
- Annotating horizontal bars in each Facet of a catplot is similar to this answer annotating vertical bars, except .get_width is used instead of .get_height, when creating custom labels.
It's easier to use pandas.crosstab to calculate percentages.
- Specify normalize='index' for each group in the 'Survey' column.
- Specify normalize='all' to normalize over all values.
Excluding creating df, this reduces your code from 27 lines to 6 lines.
Tested in python 3.11, pandas 1.5.2, matplotlib 3.6.2, seaborn 0.12.1

# beginning with the initial dataframe
df = pd.DataFrame(data)

# calculate the percentages for each group in the survey column
percent = pd.crosstab(df.survey, df.level, normalize='index').mul(100).melt(ignore_index=False).reset_index()

# plot the bars
g = sns.catplot(kind='bar', data=percent, x='value', y='level', col='survey', aspect=1.3)

# iterate through each Facet (axes)
for ax in g.axes.ravel():
    # iterate through the containers (bars)
    for c in ax.containers:
        # customize labels to not include values below a specific number
        labels = [f'{w:0.1f}%' if (w := v.get_width()) > 0 else '' for v in c]
        # add the labels
        ax.bar_label(c, labels=labels)

If there are no 0 length bars, then there's no need to specify custom labels for .bar_label. Instead use the fmt= parameter.

# iterate through each Facet (axes)
for ax in g.axes.ravel():
    # iterate through the containers
    for c in ax.containers:
        # add the labels
        ax.bar_label(c, fmt='%0.1f%%')

`percent`

     survey   level  value
0  baseline    high   30.0
1   endline    high   40.0
2  baseline     low   40.0
3   endline     low   30.0
4  baseline  medium   30.0
5   endline  medium   30.0

Alternatively, plot `df` directly

df = pd.DataFrame(data)

percent = pd.crosstab(df.survey, df.level, normalize='index').mul(100)

ax = percent.plot(kind='barh', figsize=(8, 6))

for c in ax.containers:
    ax.bar_label(c, fmt='%0.1f%%')
ax.margins(x=0.1)

How do I add bar labels in each catplot facet with horizontal bars?

2 Answers2

percent

Alternatively, plot df directly

`percent`

Alternatively, plot `df` directly