I am working with a deep NN which takes inputs X_0, X_1 and give outputs Y_0, Y_1. So X_0 and X_1 will be concatenated before going into some hidden layers and giving outputs Y_0, Y_1. My problem is that:
- I don’t know how to create a custom loss function, since Keras forces us to work with y_pred and y_true. I am not really sure how to get Y_0 and Y_1 from y_pred, but basically [Y_0, Y_1] = y_pred.
- How do I fit data into this kind of model with multiple inputs, multiple outputs and custom loss function ?
Thanks very much !