Hello,
One Model consists of two components A and B, and B takes input from A.
B is a feature extraction technique and A produces the final output.
Both A and B have a feed-forward neural network structure. Training has been done using computational graph methodology.
I want to implement a new training methodology for A using OS-ELM Algo rather than Feed Forward Neural Network which doesn’t require iterative training but B needs iterative training.
So How to implement different training methodologies under the same session of Tensorflow training for A and B, As B can’t be trained independently as it is an intermediate step that doesn’t have any of its own data for optimization?