Visualize CV scores

plot_utils.visualize_cv_scores(fig=None, ax=None, dpi=100, n_folds=5, cv_scores=None, box_height=0.6, box_width=0.9, gap_frac=0.05, metric_name='AUC', avg_cv_score=None, no_holdout_set=False, holdout_score=None, fontsize=9, flip_yaxis=True)[source]

Visualize K-fold cross-validation scores as well as hold-out set performance in an intuitive way.

Parameters:

  • fig (matplotlib.figure.Figure or None) – Figure object. If None, a new figure will be created.

  • ax (matplotlib.axes._subplots.AxesSubplot or None) – Axes object. If None, a new axes will be created.

  • dpi (float) – Figure resolution. The dpi of fig (if not None) will override this parameter.

  • n_folds (int) – Number of CV folds.

  • cv_scores (list<float> or None) – The validation score of each fold. If None, no scores will be shown on the small boxes.

  • box_height (float) – The height of the the small box, in inches.

  • box_width (float) – The width of the small box, in inches.

  • gap_frac (float) – How much gap should there be between each small box.

  • metric_name (str) – The name of the metric to be shown in the figure.

  • avg_cv_score (float or None) – The average cross-validation score. If None (recommended), it will be calculated by numpy.mean(cv_scores).

  • no_holdout_set (bool) – If False, the hold-out data set will be visualized alongside the training data set. This parameter supersedes holdout_score.

  • holdout_score (float or None) – The performance on the hold-out data set. If no_holdout_set is True, this parameter has no effect.

  • fontsize (float) – The font size of all the texts.

  • flip_yaxis (bool) – If True, everything will be flipped upside down. This parameter is for diagnosis and and debugging purpose only. It is recommended to leave it as True.

Returns:

  • fig (matplotlib.figure.Figure) – The figure object being created or being passed into this function.

  • ax (matplotlib.axes._subplots.AxesSubplot) – The axes object being created or being passed into this function.