Skip to content
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.

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

Log sigmoid #3719

Closed
vladfi1 opened this issue Aug 10, 2016 · 2 comments
Closed

Log sigmoid #3719

vladfi1 opened this issue Aug 10, 2016 · 2 comments
Assignees
Labels
type:feature Feature requests

Comments

@vladfi1
Copy link
Contributor

vladfi1 commented Aug 10, 2016

It would be nice to have a numerically stable log_sigmoid similar to the existing log_softmax.

@girving
Copy link
Contributor

girving commented Aug 10, 2016

Do you mean -tf.softplus?

@girving girving closed this as completed Aug 10, 2016
@girving
Copy link
Contributor

girving commented Apr 18, 2017

I changed my mind, after accidentally writing log_sigmoid(x) = -tf.nn.softplus(x) instead of the correct -tf.nn.softplus(-x). We should have tf.log_sigmoid, though it should be implemented as pure Python.

@girving girving reopened this Apr 18, 2017
@girving girving added stat:contribution welcome Status - Contributions welcome type:feature Feature requests labels Apr 18, 2017
@girving girving self-assigned this Apr 19, 2017
@girving girving removed the stat:contribution welcome Status - Contributions welcome label Apr 19, 2017
caisq pushed a commit to caisq/tensorflow that referenced this issue Apr 26, 2017
This is a numerically stable version of tf.log(tf.sigmoid(x)).  It's just
-tf.nn.softplus(-x), but it's easy to add and the identity is easy to mistype.

RELNOTES: Add tf.log_sigmoid(x) = tf.log(tf.sigmoid(x)) = -tf.nn.softplus(-x).

Fixes tensorflow#3719.
Change: 154308666
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
type:feature Feature requests
Projects
None yet
Development

No branches or pull requests

2 participants