Seaborn legend is cut off

[JanPalasek]

I'm using seaborn of version 0.13.0.

import seaborn as sns
import seaborn.objects as so
import matplotlib.pyplot as plt
import pandas as pd

# Load the Titanic dataset
dataset = sns.load_dataset('titanic')

data = dataset.dropna(subset=["age", "fare"])
data["age_bin"] = pd.cut(data["age"], bins=[0, 10, 20, 40, 60, 70], labels=["0-10", "11-20", "21-40", "41-60", "61-70"])
avg_fares = data.groupby(['age_bin', 'sex'])["fare"].mean().reset_index()
avg_fares["text"] = avg_fares.apply(lambda x: f"{x['fare']:.0f}" if x["fare"] > 30 and x["sex"] == "male" else "", axis=1)


fig, ax = plt.subplots()
# additional mpl settings like hiding spines
ax.grid(visible=False)
(
    so.Plot(data=avg_fares, x="age_bin", y="fare", color="sex")
    .add(so.Line(marker="o"))
    .add(so.Text(halign="left", offset=6), text="text")
    .label(x="Age", y=r"$\overline{\text{Fare}}$", color="Sex", title="Average Male Fare for given Age Group")
    .scale(color={"female": "lightgray", "male": "blue"})
    .on(ax)
    .show()
)
fig.show()

The output looks like this:

How can I persuade seaborn (or matplotlib) to include the entire legend? Also I would like the legend to be outside of the plot.

[thuiop]

Yes, matplotlib is a bit painful with figure legends. For me the easiest is to attach it to the ax directly ; see my solution in #2994.