Open shesteki opened 5 years ago
Hi, How can I create a custom layer in matconvnet with trainable weights?
i.e. in my layer, get an image, X, and compute af(bx), so that, f is a function and a,b are trainable parameters.
could anybody help me? thanks a lot.
Hi, I think you can use Autonn wrapper : https://github.com/vlfeat/autonn/blob/master/doc/TUTORIAL.md
Hi, How can I create a custom layer in matconvnet with trainable weights?
i.e. in my layer, get an image, X, and compute af(bx), so that, f is a function and a,b are trainable parameters.
could anybody help me? thanks a lot.