skopt.plots
.plot_evaluations#
- skopt.plots.plot_evaluations(result, bins=20, dimensions=None, plot_dims=None, size=2, wspace=0.35, hspace=0.35, cmap='viridis', ax=None)[source][source]#
Visualize the order in which points were sampled during optimization.
This creates a 2-d matrix plot where the diagonal plots are histograms that show the distribution of samples for each search-space dimension.
The plots below the diagonal are scatter-plots of the samples for all combinations of search-space dimensions.
The order in which samples were evaluated is encoded in each point’s color.
A red star shows the best found parameters.
- Parameters:
- result
OptimizeResult
The optimization results from calling e.g.
gp_minimize()
.- binsint, bins=20
Number of bins to use for histograms on the diagonal.
- dimensionslist of str, default=None
Labels of the dimension variables.
None
defaults tospace.dimensions[i].name
, or if alsoNone
to['X_0', 'X_1', ..]
.- plot_dimslist of str and int, default=None
List of dimension names or dimension indices from the search-space dimensions to be included in the plot. If
None
then use all dimensions except constant ones from the search-space.- sizefloat, default=2
Height (in inches) of each facet.
- wspacefloat, default=0.35
The width of the padding between subplots, as a fraction of the average Axes width. Ignored if
ax
is provided.- hspacefloat, default=0.35
The height of the padding between subplots, as a fraction of the average Axes height. Ignored if
ax
is provided.- sizefloat, default=2
Height (in inches) of each facet.
- cmap: str or Colormap, default = ‘viridis’
Color map for scatter plots. Passed directly to
plt.scatter()
- ax: `Matplotlib.Axes`, default= None
An axis object in which to plot the dependence plot.
- result
- Returns:
- ax
Matplotlib.Axes
Matplotlib axis the plto was drawn in
- ax