Vectorization of aima.core.learning.neural code
SayaliPatkar opened this issue · comments
Currently, the code is hardwired to work with a neural network with a single hidden layer. Complete vectorization of code is possible where the user can specify the number of hidden layers + hidden units. and forward and backpropagation can run in the loop according to user-specified layer + unit values. If mentors approve, I would like to work on it.
@SayaliPatkar I think you should include this in your proposal.