A few transfer calls to enable cuda #7
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
No idea how this works...just to build confidence in my changes. But I also added stuff related to #8 which was like a 2 line change, and implementing log normal dropout... |
|
can you resolve the merge conflicts? I'll merge it then. Thanks! |
|
aaand done! hopefully it's fine ...not quite sure if I removed the dividing by sigma when i made this pull request so it might still be there FYI |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I noticed your code would error when switching to use_cuda =True, so here's a few changes just to let it run. Tried it on a few of your environments that use those functions and seems to be fine across them.
More long term, the cuda calls need to be optimized because for training the policy it's faster to run it on CPUs at the moment from running it. You're probably familiar with this and focused on the more important parts, but putting things on cuda is expensive, so the less you transfer things between devices the better.