(DINOv2) Implement positional encoding interpolation #343
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
Laurent2916
force-pushed
the
pr/dinov2-pos_emb_interpolation
branch
3 times, most recently
from
00bd5fa
to
5bc84f2
Compare
Laurent2916
force-pushed
the
pr/dinov2-pos_emb_interpolation
branch
from
5bc84f2
to
f720e4f
Compare
deltheil
approved these changes
Mar 29, 2024
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Outdated suggestions cannot be applied.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 PR allows DINOv2 models to compute embeddings on tensors whose "resolution" is different from the default 518x518px resolution.
This implementation doesn't use the
scale_factorargument oftorch.nn.interpolate, as per facebookresearch/dinov2#378.