Open arilwan opened 1 year ago
I am using FastTreeSHAP to help explained my random forest model prediction having over 40K samples and 24 features.
FastTreeSHAP
explainer = fasttreeshap.TreeExplainer(model, algorithm='auto' ,n_jobs=-1) shap_values = explainer(X_test).values shap_values.shape (40682, 24, 5) # plotting fasttreeshap.summary_plot(shap_values, X_test, plot_type = 'bar')
The output figure is (showing top-5 features interaction):
I expected output figure to look similar to below image (FastTreeSHAP docs).
Is there any change in the summary_plot() method?
summary_plot()
I am using
FastTreeSHAP
to help explained my random forest model prediction having over 40K samples and 24 features.The output figure is (showing top-5 features interaction):
I expected output figure to look similar to below image (FastTreeSHAP docs).
Is there any change in the
summary_plot()
method?