Experiment 3

target = (
   transformed_0 * transformed_1 * transformed_2 * transformed_3
   + np.random.normal(loc=0, scale=0.2, size=n_samples)
 )

Interaction Analysis

feature_0 - feature_1

Prediction plot:

Prediction values between feature 0 and feature 1

Function plot:

Actual interaction values between feature 0 and feature 1

treemind plot:

treemind interaction values between feature 0 and feature 1

SHAP plot:

SHAP interaction values between feature 0 and feature 1

feature_0 - feature_2

Prediction plot:

Prediction values between feature 0 and feature 2

Function plot:

Actual interaction values between feature 0 and feature 2

treemind plot:

treemind interaction values between feature 0 and feature 2

SHAP plot:

SHAP interaction values between feature 0 and feature 2

feature_0 - feature_3

Prediction plot:

Prediction values between feature 0 and feature 3

Function plot:

Actual interaction values between feature 0 and feature 3

treemind plot:

treemind interaction values between feature 0 and feature 3

SHAP plot:

SHAP interaction values between feature 0 and feature 3

feature_1 - feature_2

Prediction plot:

Prediction values between feature 1 and feature 2

Function plot:

Actual interaction values between feature 1 and feature 2

treemind plot:

treemind interaction values between feature 1 and feature 2

SHAP plot:

SHAP interaction values between feature 1 and feature 2

feature_1 - feature_3

Prediction plot:

Prediction values between feature 1 and feature 3

Function plot:

Actual interaction values between feature 1 and feature 3

treemind plot:

treemind interaction values between feature 1 and feature 3

SHAP plot:

SHAP interaction values between feature 1 and feature 3

feature_2 - feature_3

Prediction plot:

Prediction values between feature 2 and feature 3

Function plot:

Actual interaction values between feature 2 and feature 3

treemind plot:

treemind interaction values between feature 2 and feature 3

SHAP plot:

SHAP interaction values between feature 2 and feature 3