Plot policy-tree decision and projection panels

Builds the standard two-panel policy-tree display: the branching decision tree in panel A, with the projection plot of evaluation points below in panel B.

Usage

margot_plot_policy_tree_panels(
  result_object,
  model_name,
  max_depth = 2L,
  original_df = NULL,
  label_mapping = NULL,
  show_leaf_metrics = TRUE,
  leaf_metrics = NULL,
  layout = list(heights = c(1, 2)),
  annotation = list(tag_levels = "A"),
  projection_args = list(),
  decision_tree_args = list()
)

Arguments

result_object

A list returned by margot_causal_forest() or a compatible policy-tree workflow object.

model_name

Character scalar naming the model to plot.

max_depth

Integer, usually 1 or 2.

original_df

Optional data frame with original-scale variables.

label_mapping

Optional named list used for display labels.

show_leaf_metrics

Logical. If TRUE, the decision tree panel includes leaf T-C contrasts and sample shares.

leaf_metrics

Optional output from margot_policy_leaf_summary() or margot_table_policy_tree().

layout

List passed to margot_plot_policy_combo() for panel heights.

annotation

List passed to margot_plot_policy_combo() for tags.

projection_args

Optional list of arguments passed to margot_plot_policy_projection().

decision_tree_args

Optional list of arguments passed to margot_plot_policy_decision_tree().

Value

A list with decision_tree, projection, and combined_plot.