Extended neva model to support llava_next #11390
Closed
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
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
| @@ -27,7 +27,7 @@ | |||
|
|
|||
| if TYPE_CHECKING: | |||
| from transformers import LlavaConfig as HFLlavaConfig | |||
| from transformers import LlavaForConditionalGeneration | |||
| from transformers import LlavaForConditionalGeneration, LlavaNextForConditionalGeneration | |||
Check notice
Code scanning / CodeQL
Unused import
|
beep boop 🤖: 🚨 The following files must be fixed before merge! Your code was analyzed with PyLint. The following annotations have been identified: Thank you for improving NeMo's documentation! |
|
beep boop 🤖: 🙏 The following files have warnings. In case you are familiar with these, please try helping us to improve the code base. Your code was analyzed with PyLint. The following annotations have been identified: Thank you for improving NeMo's documentation! |
| @@ -31,7 +39,9 @@ | |||
| "NevaConfig", | |||
| "NevaModel", | |||
| "LlavaConfig", | |||
| "Llava15Config7B", | |||
| "LlavaNextModel" "Llava15Config7B", | |||
Check failure
Code scanning / CodeQL
Explicit export is not defined
| @@ -31,7 +39,9 @@ | |||
| "NevaConfig", | |||
| "NevaModel", | |||
| "LlavaConfig", | |||
| "Llava15Config7B", | |||
| "LlavaNextModel" "Llava15Config7B", | |||
Check warning
Code scanning / CodeQL
Implicit string concatenation in a list
yaoyu-33
reviewed
Nov 25, 2024
| image_newline=self.image_newline, | ||
| ) | ||
| combined_embeddings, attention_mask, position_ids, final_labels, final_input_ids, final_loss_mask = ( | ||
| merge_input_ids_with_image_features( |
There was a problem hiding this comment.
I don't like this design here, reason -
this method is very similar function as self._preprocess_data. In self._preprocess_data we do more things to support sp and cp. It will be good if we can re-use the same method. Can you tell me how these 2 methods are different?
yaoyu-33
reviewed
Nov 25, 2024
| @@ -455,6 +459,7 @@ def forward( | |||
| self, | |||
| input_ids: torch.Tensor, | |||
| position_ids: torch.Tensor, | |||
| image_sizes: List[torch.Tensor] = None, | |||
There was a problem hiding this comment.
if we are going to change the signiture, it would be better we are using 2 model class? it will might break something we will use nemo / mcore llava
There was a problem hiding this comment.
It should not break anything for vanilla neva as image_sizes tensor is optional and will be used only for llava_next. The reason for using the neva model class is most of the functionality is same.
yashaswikarnati
force-pushed
the
yash/neva_model_changes
branch
from
8c69d24
to
8f779ba
Compare
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 ?
Extended neva model to support llava_next
Collection:
VLM
Changelog
-NeVA fwd function changes to handle image tiling
GitHub Actions CI
The Jenkins CI system has been replaced by GitHub Actions self-hosted runners.
The GitHub Actions CI will run automatically when the "Run CICD" label is added to the PR.
To re-run CI remove and add the label again.
To run CI on an untrusted fork, a NeMo user with write access must first click "Approve and run".
Before your PR is "Ready for review"
Pre checks:
PR Type:
If you haven't finished some of the above items you can still open "Draft" PR.
Who can review?
Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.
Additional Information