ValueError when applying mlxtend Sequential Feature Selection on wrapped Keras neural net

Question

I\'m hoping to apply Sequential Feature Selection from mlxtend to select the best subset of input feautures for my classification dense neural network.

However, I\'m stuc