treemind.plot

treemind.plot.feature_plot(result, col, figsize=(12.0, 8.0), show_std=False, show_range=False, xticks_n=10, yticks_n=10, ticks_fontsize=10.0, title_fontsize=16.0, label_fontsizes=14.0, title=None, xlabel=None, ylabel=None)

Plot a single feature’s contribution profile, handling either

  • a continuous feature stored as split-point intervals (<feature>_lb / <feature>_ub), or

  • a categorical feature stored as one identifier column.

The plot shows the mean impact (value) and can optionally display one or both of

  • ± std shaded bands, and

  • a count / frequency distribution (kde for continuous, line plot for categorical).

New: Supports multiclass problems. If a ‘class’ column is present, separate plots will be generated for each class.

Parameters:

  • result (Result) – The result object containing the feature data.

  • col (int) – The index of the feature to plot.

  • figsize ((float, float), default (12, 8)) – Width × height of the figure (in inches).

  • show_std (bool, default False) – If True, plot ± std shaded regions.

  • show_range (bool, default True) – If True, draw the count / frequency distribution: a KDE curve for continuous data or a line-and-fill plot for categorical data.

  • xticks_n (int, default 10) – Desired number of tick marks on the x- and y-axes respectively.

  • yticks_n (int, default 10) – Desired number of tick marks on the x- and y-axes respectively.

  • ticks_fontsize (float, default 10.0) – Font size for tick labels.

  • title_fontsize (float, default 16.0) – Font size for the plot title (if drawn).

  • label_fontsizes (float, default 14.0) – Font size for x- and y-axis labels and legend text.

  • title (str or None, default None) – Custom figure title. If None, a title of the form "Contribution of <feature> - Class X" is generated automatically.

  • xlabel (str or None, default None) – Custom axis labels. If None, they default to the feature name (for x) and "Value" (for y).

  • ylabel (str or None, default None) – Custom axis labels. If None, they default to the feature name (for x) and "Value" (for y).

Returns:

The function draws the figure(s) and shows them; nothing is returned.

Return type:

None

Notes

  • Continuous data is rendered with a step line (tree-split style). Interval endpoints of ± ∞ are compressed to the outermost finite split so the geometry remains well defined.

  • Categorical data is rendered with vertical bars originating from zero; categories appear in the order they occur in df.

  • Multiclass data generates separate plots for each class, with class information appended to the plot title.

Example Usage

Below is an example of how to use the feature_plot function:

from treemind import Explainer
from treemind.plot import feature_plot

# Assume 'model' is a trained tree model

# Create an instance of the Explainer
explainer = Explainer(model)

# Analyze the specified feature by its index
result = explainer.explain(degree=1)

# Plot the feature using a line plot
feature_plot(result, 22)

Output

feature_plot example
treemind.plot.interaction_plot(result, cols, figsize=(10.0, 8.0), axis_ticks_n=10, ticks_fontsize=10, title_fontsize=16, label_fontsizes=14, title=None, xlabel=None, ylabel=None, color_bar_label=None)

Plot a heat-map of interaction strengths for two features that may be continuous, categorical, or one of each. If a class column is provided, creates separate plots for each class.

A continuous feature must appear as column pair <name>_lb + <name>_ub (interval bounds); a categorical feature appears as a single column <name>. The DataFrame must also include value (mean impact), std (standard deviation) and count (bin size).

Parameters:

  • result (Result) – Result object containing interaction statistics.

  • cols (tuple of int or list of int) – Indices of the two features to plot.

  • figsize ((float, float), default (10, 8)) – Width × height of the figure.

  • axis_ticks_n (int, default 10) – Number of ticks to show on a continuous axis.

  • ticks_fontsize (float, default 10) – Font size for tick labels.

  • title_fontsize (float, default 16) – Font size for the title.

  • label_fontsizes (float, default 14) – Font size for axis- and colour-bar labels.

  • title (str or None, default None) – Optional figure title.

  • xlabel (str or None, default None) – Optional axis labels (fall back to feature names).

  • ylabel (str or None, default None) – Optional axis labels (fall back to feature names).

  • color_bar_label (str or None, default None) – Label for the colour-bar (“Impact” if None).

  • class_column (str or None, default None) – Name of the class column. If provided, creates separate plots for each class.

Returns:

Displays the figure(s).

Return type:

None

Example Usage

Below is an example of how to use the interaction_plot function:

from treemind import Explainer
from treemind.plot import interaction_plot

# Assume 'model' is a trained tree model

# Create an instance of the Explainer
explainer = Explainer(model)

# Analyze the specified feature by its index
result = explainer.explain(degree=2)

# Plot the feature using a line plot
interaction_plot(result, (22, 21))

Output

interaction_plot example
treemind.plot.interaction_scatter_plot(X, result, cols, figsize=(10.0, 8.0), ticks_fontsize=10.0, title_fontsize=16.0, label_fontsizes=14.0, title=None, xlabel=None, ylabel=None, color_bar_label=None)

Creates a scatter plot of feature values with colors representing interaction values. If ‘class’ column exists in df, creates separate plots for each class.

Parameters:

  • X (pd.DataFrame) – Input data containing feature values

  • result (Result) – Result object containing interaction statistics.

  • cols (tuple of int or list of int) – Indices of the two features to plot.

  • figsize (tuple of float, default (10.0, 8.0)) – Width and height of the plot in inches.

  • ticks_fontsize (float, default 10.0) – Font size for axis tick labels,

  • title_fontsize (float, default 16.) – Font size for plot title, by

  • title (str, optional, default None) – The title displayed at the top of the plot. If None, no title is shown.

  • xlabel (str, optional, default None) – Label for the x-axis. If None, it will default to the feature name.

  • ylabel (str, optional, default None) – Label for the y-axis. If None, it will default to the feature name.

  • color_bar_label (str, optional, default None) – Colorbar label, If None, it will default to “Impact”.

  • label_fontsizes (float)

Return type:

None

Example Usage

Below is an example of how to use the interaction_scatter_plot function:

from treemind import Explainer
from treemind.plot import interaction_scatter_plot

# Assume 'model' is a trained tree model

# Create an instance of the Explainer
explainer = Explainer(model)

# Analyze the specified feature by its index
result = explainer.explain(degree=2)

# Plot the feature using a line plot
interaction_scatter_plot(X, result, (22, 21))

Output

interaction_scatter_plot example
treemind.plot.importance_plot(result, combine_classes=False, figsize=(10.0, 6.0), ticks_fontsize=10.0, title_fontsize=16.0, label_fontsize=14.0, cmap='Reds', *, title=None)

Visualise feature-importance scores with the same visual style as interaction_plot and feature_plot.

Parameters:

  • result (Result) – The result object containing the feature importance data.

  • combine_classes (bool, default False) – If True, combines class-wise importances into a single DataFrame. Only applicable for multi-class models.

  • figsize ((float, float), default (10, 6)) – Figure size in inches.

  • ticks_fontsize (float) – Font sizes for ticks, title and axis/legend labels.

  • title_fontsize (float) – Font sizes for ticks, title and axis/legend labels.

  • label_fontsize (float) – Font sizes for ticks, title and axis/legend labels.

  • cmap (str, default "coolwarm") – Diverging colour-map centred at zero for the heat-map.

  • title (str or None) – Custom figure title (string is prepended with class info when multiclass). If None, an automatic title is generated.

Return type:

None

Example Usage

Below is an example of how to use the importance_plot function:

from treemind import Explainer
from treemind.plot import importance_plot

# Assume 'model' is a trained tree model

# Create an instance of the Explainer
explainer = Explainer(model)

# Analyze the specified feature by its index
result = explainer.explain(degree=1)

# Plot the feature using a line plot
importance_plot(result)

Output

interaction_scatter_plot example