Skip to content

Conversation

@Smerity
Copy link
Contributor

@Smerity Smerity commented Mar 3, 2017

Minimal change: essentially removing code that now duplicates code included in PyTorch's built-ins. This also fixes a potential bug that was reported regarding reaching the same perplexity numbers reported in the code.
Fixes #95.

@Smerity Smerity changed the title Replace clip_gradient with torch's newly included clip_grad_norm (fixes #95) Replace clip_gradient with torch's newly included clip_grad_norm Mar 3, 2017
@apaszke apaszke merged commit 0940311 into pytorch:master Mar 3, 2017
@apaszke
Copy link
Contributor

apaszke commented Mar 3, 2017

Thanks!

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

LSTM language model baseline gap

2 participants