tabensemb.trainer.Trainer.plot_loss#
method
- Trainer.plot_loss(program: str, model_name: str, ax=None, train_val: str = 'both', restored_epoch_mark: bool = True, restored_epoch_mark_if_last: bool = False, legend: bool = True, clr: Iterable | None = None, plot_kwargs: Dict | None = None, scatter_kwargs: Dict | None = None, legend_kwargs: Dict | None = None, figure_kwargs: Dict | None = None, savefig_kwargs: Dict | None = None, save_show_close: bool = True) Axes[source]#
Plot loss curves for a model.
- Parameters:
- program
The selected model base.
- model_name
The selected model in the model base.
- ax
matplotlib.axes.Axes- train_val
“train” to plot training loss only. “val” to plot validation loss only. “both” to plot both of them.
- restored_epoch_mark
Plot the best epoch from where the model is restored after training.
- restored_epoch_mark_if_last
Plot the best epoch when it is the last epoch.
- legend
Show legends.
- clr
A seaborn color palette or an Iterable of colors. For example seaborn.color_palette(“deep”).
- plot_kwargs
Arguments for
plt.plot- scatter_kwargs
Arguments for
plt.scatter(used to plot the restored epoch).- legend_kwargs
Arguments for
plt.legend.- figure_kwargs
Arguments for
plt.figure.- savefig_kwargs
Arguments for
plt.savefig- save_show_close
Whether to save, show (in the notebook), and close the figure if
axis not given.
- Returns:
- matplotlib.axes.Axes