CODEWITHSUNDEEP
pythonmatplotlibseaborn
bzwanenburg
bzwanenburg

Reputation: 85

Seaborn plots, legend overlaps figure

Using Seaborn, my legend keeps overlapping the data, no matter what dataframe I use, or whether I use pairplots or jointgrids. I know I can work around this by removing Seaborn's legend and making a custom legend, however, that's not the "cleanest" route. How can I get Seaborn to create non-overlapping legends ?

Here some code:

g = sns.pairplot(df, kind="reg", plot_kws={"marker": "+"}, hue="experiment", palette="Set2", x_vars=["alpha [%]", "shelter [%]", "beta [%]"], y_vars=["final [%]"])

plt.show()

(btw I'm on Mac OS, Pycharm,enter image description here Python 3.6seaborn 0.10.0 and matplotlib 3.3.3)

Upvotes: 1

Views: 832

Answers (1)

kir0ul
kir0ul

Reputation: 69

It seems you can not do it with pairplot. In the docs they say:

This is a high-level interface for PairGrid that is intended to make it easy to draw a few common styles. You should use PairGrid directly if you need more flexibility.

Taking this PairGrid example from the docs, you can pass the loc parameter to the add_legend() method.

g = sns.PairGrid(penguins, hue="species")
g.map_diag(sns.histplot)
g.map_offdiag(sns.scatterplot)
g.add_legend(loc=(0.9,0.2))  # or g.add_legend(loc="upper right");
plt.show()

The arguments you can pass to the loc parameter are listed in the Matplotlib docs.

Upvotes: 1

Related Questions