Adding details to NTXentLoss documentation #650
Merged
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Outdated suggestions cannot be applied.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.
Hello! I use this repo so much I figured I ought to contribute where I can. I use
NTXentLoss
quite often and figured I could take a try at adding some documentation and address Issue #608.I drew largely from @KevinMusgrave's comment to create an example of how embeddings are handled by the loss function. I realize much of the information I added is probably not specific enough to this loss function, so feel free to trim and cut to make it more concise.
(You can also scrap this PR entirely if it isn't what you had in mind for #608! I figured so many people are interested in temperature normalized cross-entropy loss that some additional documentation would benefit us all 😄)