Dropout #182
Open
Dropout #182
Conversation
This file contains hidden or 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
| local state = cudnn.dropout_rng_states[dev] | ||
| errcheck('cudnnSetDropoutDescriptor', self.dropDesc[0], | ||
| cudnn.getHandle(), self.p, | ||
| state:data(), state:nElement(), torch.seed()) |
There was a problem hiding this comment.
this would reset the seed of torch RNG - this is probably not something you want to do, as it can introduce non-desirable non-determinism. Better to use torch.random() here?
|
cudnn.Dropout:setp() will be broken (for it to work, cudnn.Dropout:setp() should call setDropoutDescriptor using NULL as state argument if states are initialized already). (and thanks for these bindings!) |
|
There is a potential problem with several dropout layers sharing the states and running concurrently in the different streams, but its hard to make this scenario completely safe. |
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Outdated suggestions cannot be applied.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.
cudnn.Dropout should take about a ton less memory because it stores masks in binary format and faster as well, thanks to @ngimel for explaining this.
can someone review please?
Btw, R5 is released, it's time to make it master @soumith ?