Experiment 1
target = (
0.4 * transformed_0
- 0.6 * transformed_1
+ 0.3 * transformed_2
+ 0.5 * transformed_3
- 0.4 * transformed_4
+ 0.7 * transformed_5
- 0.3 * transformed_6
+ 0.5 * transformed_7
- 0.4 * transformed_8
+ 0.6 * interaction_0_2
- 0.5 * interaction_1_3
+ 0.4 * interaction_4_6
- 0.3 * interaction_5_7
+ 0.5 * interaction_6_8
+ np.random.normal(loc=0, scale=0.2, size=n_samples)
)
Feature Analysis
feature_0
Function plot:
treemind plot:
SHAP plot:
feature_1
Function plot:
treemind plot:
SHAP plot:
feature_2
Function plot:
treemind plot:
SHAP plot:
feature_3
Function plot:
treemind plot:
SHAP plot:
feature_4
Function plot:
treemind plot:
SHAP plot:
feature_5
Function plot:
treemind plot:
SHAP plot:
feature_6
Function plot:
treemind plot:
SHAP plot:
feature_7
Function plot:
treemind plot:
SHAP plot:
feature_8
Function plot:
treemind plot:
SHAP plot:
Interaction Analysis
feature_0 - feature_2
Prediction plot:
Function plot:
treemind plot:
SHAP plot:
feature_1 - feature_3
Prediction plot:
Function plot:
treemind plot:
SHAP plot:
feature_4 - feature_6
Prediction plot:
Function plot:
treemind plot:
SHAP plot:
feature_5 - feature_7
Prediction plot:
Function plot:
treemind plot:
SHAP plot:
feature_6 - feature_8
Prediction plot:
Function plot:
treemind plot:
SHAP plot: