[BLIP-2] Improve conversion script #24854
Merged
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
|
The documentation is not available anymore as the PR was closed or merged. |
NielsRogge
force-pushed
the
improve_blip2
branch
from
1ae55de to
14f9f2f
Compare
NielsRogge
force-pushed
the
improve_blip2
branch
from
0c06392 to
f0d6e3d
Compare
Closed
4 tasks
amyeroberts
reviewed
Jul 31, 2023
There was a problem hiding this comment.
Thanks for the work updating this! In particular getting the comparison with the original model. It makes the script a lot cleaner.
Two main comments:
- Does this mean that the currently converted checkpoints on the hub need to be updated?
- The doc tests for the model should be added to documentation_tests.txt
src/transformers/models/blip_2/convert_blip_2_original_to_pytorch.py
Outdated
Show resolved
Hide resolved
src/transformers/models/blip_2/convert_blip_2_original_to_pytorch.py
Outdated
Show resolved
Hide resolved
|
This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread. Please note that issues that do not follow the contributing guidelines are likely to be ignored. |
amyeroberts
approved these changes
Sep 13, 2023
src/transformers/models/blip_2/convert_blip_2_original_to_pytorch.py
Outdated
Show resolved
Hide resolved
src/transformers/models/blip_2/convert_blip_2_original_to_pytorch.py
Outdated
Show resolved
Hide resolved
|
Feel free to merge :) |
parambharat
pushed a commit
to parambharat/transformers
that referenced
this pull request
Sep 26, 2023
* Improve conversion script * Add int8 code example * Update tip * Fix code * Fix code snippet * Add nucleus sampling * More improvements * Address comments * Address comments
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.
What does this PR do?
When investigating an issue reported here, I've reran and improved BLIP-2's conversion script (based on InstructBLIP).
It's important to compare apples-to-apples, so I had to fork the LAVIS repo and make sure the original model is also run in float32.