[Python] Which shap explainer should be used for multimodal input (image data and tabular...

I want to try explaining the decisions of a model in pytorch with 2 inputs of image and tabular data like it is used in this tutorial (https://rosenfelder.ai/multi-input-neural-network-pytorch/)

If I do want to interpret such a model with shapley values which explainer e.g. of the shap package would be the right one? Or do I have to modify the algorithm?

Continue reading...