Lazy initialization of buffers #617
Merged
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
SpatialConvolution.lua
Outdated
| self.finput = torch.Tensor() | ||
| self.fgradInput = torch.Tensor() | ||
| self.finput = self.finput and self.weight.new() | ||
| self.fgradInput = self.fgradInput and self.weight.new() |
Contributor
There was a problem hiding this comment.
CPU and GPU version of SpatialConvolutionMM uses differently fgradInput (GPU assumes that fgradInput contains all ones). That's why it was being newly allocating this tensor. I think you should at least resize it to zero. See #211
Contributor
There was a problem hiding this comment.
I actually think you should do it for finput as well (as it was before), because converting an already forwarded CPU model to GPU would create a huge (and unnecessary) finput buffer.
Member
|
Thanks! |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
For backend switching mostly, see also the discussion here #526