I’m modifying Resnet’s code so that each layer can take another set of inputs.
Resnet has layers which are a Sequential object.
This is the code I’m using to run a tuple of inputs (image,aux_input):
def run_layer(some_layer,x_aux): x,aux_info=x_aux for i in range(len(some_layer)): x=some_layer[i].forward((x,aux_info)) return x
Assume that each layer does return some kind of an image encodings which have nothing to do with aux_info.
I’m wondering if my code is correct and if there’s a better way to code it -I have a feeling there’s something wrong I’m doing so that’s why this post.