New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Softplus with parameters #5939

Open

nishnik opened this issue May 12, 2017 · 3 comments

Labels

Contribution Welcome

nishnik commented May 12, 2017

Should I go on implementing the parametric version of Softplus?

Current implementation: nnet/sigm.py#L29

Reference to parametric version: PyTorch docs

nouiz added the Contribution Welcome label

Member

nouiz commented May 12, 2017 via email •

edited

It would be a good contribution to Theano. The current softplus is down in that file: https://github.com/Theano/Theano/blob/master/theano/tensor/nnet/sigm.py#L319 Here is the doc with example on how to do that. It point to all change needed. http://deeplearning.net/software/theano/extending/other_ops.html#scalar-elemwise-reduction-ops

Author

nishnik commented Jul 30, 2017

Hey! @nouiz I tried to make changes here.

I do not know how to carry forward, its giving errors in grad. Could you please help?

Member

nouiz commented Aug 3, 2017

Make a PR out of it. It make discussion on the code much easier then regular commit.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment