Can we take the average of Shapley values across multiple models?

Question

I am seeking to interpret my Keras Tensorflow model in Python so I used the SHAP KernelExplainer function to calculate Shapley values, which are indicative of the contributi