Visualization of Optimization Results

If you would like to plot the optimization process, you can use one of the following built-in functions from the sklearn_nature_inspired_algorithms.helpers package.

score_by_generation_lineplot

sklearn_nature_inspired_algorithms.helpers.score_by_generation_lineplot plots the score (min, max, mean, or median) of each algorithm run by generation using a line plot. The score, or “fitness” in nature-inspired algorithms, is set by the provided estimator. Estimators have a score method that provides a default evaluation criterion for the problem they are designed to solve (reference). You can set your own score by providing the scoring parameter to NatureInspiredSearchCV.

score_by_generation_lineplot(nia_search, metric='max', ax=None, ylim=None)

Parameters

nia_search: NatureInspiredSearchCV object
metric: str, default=’max’. The metric to plot. Possible values are min, max, mean, and median.
ax: matplotlib Axes, default=None. The axes object to plot on. If not provided, one will be created.
ylim: tuple, default=None. Only applicable if the ax parameter is None. Sets the boundaries of the Y axis.

Returns

ax: matplotlib Axes

Returns the Axes object with the plot drawn on it.

score_by_generation_violinplot

sklearn_nature_inspired_algorithms.helpers.score_by_generation_violinplot plots the score of a selected run by generation using violin plots.

score_by_generation_violinplot(nia_search, run=0, ax=None, ylim=None)

Parameters

nia_search: NatureInspiredSearchCV object
run: int, default=0. The run to plot.
ax: matplotlib Axes, default=None. The axes object to plot on. If not provided, one will be created.
ylim: tuple, default=None. Only applicable if the ax parameter is None. Sets the boundaries of the Y axis.

Returns

ax: matplotlib Axes

Returns the Axes object with the plot drawn on it.