xgboost plot importance figure size

Question

How can I change the figure size of xgboost's plot importance function?

Trying to pass a figsize=(10,20) fails with the exception of unknown attribute.

Answer 1

You can pass an axis in the ax argument in plot_importance(). For instance, use this wrapper:

def my_plot_importance(booster, figsize, **kwargs): 
    from matplotlib import pyplot as plt
    from xgboost import plot_importance
    fig, ax = plt.subplots(1,1,figsize=figsize)
    return plot_importance(booster=booster, ax=ax, **kwargs)

Then use my_plot_importance() instead of plot_importance()

Answer 2

You can also set the figure size with:

from xgboost import plot_importance
import matplotlib.pyplot as plt

plt.rcParams["figure.figsize"] = (14, 7)
plot_importance(booster=booster)

Answer 3

Only add plt.rcParams["figure.figsize"] = (20,50) to your code

For example:

from xgboost import plot_importance
plt.figure(figsize=(40,20))
plot_importance(model,max_num_features=100)
plt.rcParams["figure.figsize"] = (20,100)
plt.show()

Adjust (20,100) to enlarge or reduce image size