building a neural network on pytorch and trying to get the fully connected layer to accept the dimension that comes out of a flatten layer i call before it.
so far i\
