Proposal of how we might use sparse tensors for gradients #1484
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
Dipet
requested review from
awan-10,
cli99,
conglongli,
eltonzheng,
jeffra,
minjiaz,
niumanar,
RezaYazdaniAminabadi,
samyam,
ShadenSmith and
tjruwase
as code owners
|
Thank you @Dipet for your contribution, sorry for delay on our side reviewing this. I’ll take a closer look this afternoon. |
jeffra
approved these changes
Oct 26, 2021
There was a problem hiding this comment.
This is great, thank you @Dipet for the contribution. This CSR tensor was implemented a really long time ago and I don't believe back then torch had support for this properly. Really glad to see this feature be upgraded like this.
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Outdated suggestions cannot be applied.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.
This is a simple solution to how we could use sparse gradients for the Embedding and EmbeddingBag.
When I was writing this code, I saw that your CSRTensor implementations looks like COO tensor in pytorch because pytorch CSRTensor has another API.
I created a new PR because the rebase operation went wrong in #1439