-
-
Notifications
You must be signed in to change notification settings - Fork 611
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Simplify Embedding
#2084
Merged
Merged
Simplify Embedding
#2084
Changes from all commits
Commits
Show all changes
3 commits
Select commit
Hold shift + click to select a range
File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
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
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
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Outdated suggestions cannot be applied.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.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
These could instead call
Flux.onecold
. The result will differ on e.g.[true, true, false]
, not sure we care too much either way?There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For performance in the one hot case? If it's
onecold
-compatible, then folks should useOneHotArray
for performance. At least with*
, we do the mathematically expected operation.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
For
OneHotArray
these should be identical, right? Result and performance.For a one-hot BitArray, the results will agree. I would guess that onecold is faster but haven't checked.
For a generic BitArray, I'm not sure which is mathematically expected really. I think you're saying that
*
is.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, what you wrote is what I meant re: performance. I was adding that in the one-hot bit array case, we can direct people to
OneHotArray
if their concern is performance.Yeah whenever I've come across this type of operation in papers, I see it written as
*
. There's an implicit assumption thatx
is one-hot, so maybeonecold
could be better here if it were made to error for[true, true, false]
, etc. But I think silently choosing the first "hot" index is wrong.There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes. Mixing two embedding vectors seems less wrong. But probably nobody ever hits this & it's just a way to decouple from OneHotArray types. I don't think we should document that boolean indexing is an option.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
So I think we are happy with the current implementation in the PR?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes I think so.
I see we had a very similar discussion in #1656 (comment) BTW, I forgot... but same conclusion.