[Bugfix] Keye-VL compatibility with tok_kwargs (#20058)
#20353
Conversation
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
There was a problem hiding this comment.
Summary of Changes
Hello @DarkLight1337, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!
This pull request resolves a compatibility bug within the Keye-VL model's processing logic. The primary goal is to ensure that tokenizer-specific keyword arguments (tok_kwargs) are correctly propagated to the underlying Hugging Face processor, which was causing issues with basic model tests. The changes also refine how multimodal keyword arguments (mm_kwargs) are handled before being passed to the processor.
Highlights
- Bugfix: Addresses a compatibility issue for the
Keye-VLmodel, specifically ensuringtok_kwargs(tokenizer keyword arguments) are correctly handled during processing. - Parameter Propagation: Modifies the
_call_hf_processormethod to accepttok_kwargsand correctly pass them, alongside processedmm_kwargs, to the underlying Hugging Face processor. - Multimodal Kwargs Pre-processing: Introduces a step to pre-process
mm_kwargsusingself.info._get_image_processor_kwargsbefore they are used in the final call to the Hugging Face processor.
Using Gemini Code Assist
The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.
Invoking Gemini
You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.
| Feature | Command | Description |
|---|---|---|
| Code Review | /gemini review |
Performs a code review for the current pull request in its current state. |
| Pull Request Summary | /gemini summary |
Provides a summary of the current pull request in its current state. |
| Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in issue comments and review comments. |
| Help | /gemini help |
Displays a list of available commands. |
Customization
To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.
Limitations & Feedback
Gemini Code Assist is currently in preview and may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments to provide feedback.
You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.
Footnotes
-
Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩
| mm_kwargs = self.info._get_image_processor_kwargs(**mm_kwargs) | ||
| return self.info.ctx.call_hf_processor( | ||
| self.info.get_hf_processor(**mm_kwargs), | ||
| dict(text=prompt, **mm_data), | ||
| self.info._get_image_processor_kwargs(**mm_kwargs), | ||
| dict(**mm_kwargs, **tok_kwargs), |
There was a problem hiding this comment.
To improve code clarity and avoid variable shadowing, it would be beneficial to use a new variable for the processed kwargs. This makes it explicit that you are using the processed arguments for both getting the processor and for the final processor call.
| mm_kwargs = self.info._get_image_processor_kwargs(**mm_kwargs) | |
| return self.info.ctx.call_hf_processor( | |
| self.info.get_hf_processor(**mm_kwargs), | |
| dict(text=prompt, **mm_data), | |
| self.info._get_image_processor_kwargs(**mm_kwargs), | |
| dict(**mm_kwargs, **tok_kwargs), | |
| processed_mm_kwargs = self.info._get_image_processor_kwargs(**mm_kwargs) | |
| return self.info.ctx.call_hf_processor( | |
| self.info.get_hf_processor(**processed_mm_kwargs), | |
| dict(text=prompt, **mm_data), | |
| dict(**processed_mm_kwargs, **tok_kwargs), | |
| ) |
|
👋 Hi! Thank you for contributing to the vLLM project. 💬 Join our developer Slack at https://slack.vllm.ai to discuss your PR in #pr-reviews, coordinate on features in #feat- channels, or join special interest groups in #sig- channels. Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run Once the PR is approved and ready to go, your PR reviewer(s) can run CI to test the changes comprehensively before merging. To run CI, PR reviewers can either: Add 🚀 |
…vllm-project#20353) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
…vllm-project#20353) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
…vllm-project#20353) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk> Signed-off-by: avigny <47987522+avigny@users.noreply.github.com>
…vllm-project#20353) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
…vllm-project#20353) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk> Signed-off-by: Jinzhen Lin <linjinzhen@hotmail.com>
…vllm-project#20353) Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.Purpose
Keye-VL is broken on main because it is incompatible with #20058
Fix Basic Models test
Test Plan
Test Result
(Optional) Documentation Update