Improve perceiver #14750
Merged
Improve perceiver #14750
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
LysandreJik
approved these changes
Dec 13, 2021
Member
LysandreJik
left a comment
LysandreJik
left a comment
There was a problem hiding this comment.
I think this is a very welcome change. Thank you for working on this, @NielsRogge!
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
What does this PR do?
This PR removes the need for the hard-coded
d_modelattribute of the Perceiver. Instead, one calculates the dimensionality (i.e. number of channels) of the inputs based on thenum_channelsproperty of the preprocessor.Note that
d_modelis still relevant for users 1) when embedding text usingPerceiverTextProcessor2) when no preprocessor is provided.