tabensemb.trainer.Trainer.plot_partial_err#

method

Trainer.plot_partial_err(program: str, model_name: str, feature, thres=0.8, ax=None, clr: Iterable | None = None, figure_kwargs: Dict | None = None, scatter_kwargs: Dict | None = None, hist_kwargs: Dict | None = None, savefig_kwargs: Dict | None = None, save_show_close: bool = True) Axes[source]#

Calculate prediction absolute errors on the testing dataset, and plot histograms of high-error samples and low-error samples respectively for a single feature.

Parameters:
program

The selected model base.

model_name

The selected model in the model base.

feature

The selected feature.

thres

The absolute error threshold to identify high-error samples and low-error samples.

ax

matplotlib.axes.Axes

clr

A seaborn color palette or an Iterable of colors. For example seaborn.color_palette(“deep”).

figure_kwargs

Arguments for plt.figure.

scatter_kwargs

Arguments for ax.scatter()

hist_kwargs

Arguments for ax.hist()

savefig_kwargs

Arguments for plt.savefig

save_show_close

Whether to save, show (in the notebook), and close the figure if ax is not given.

Returns:
matplotlib.axes.Axes