gemini-code-assist[bot] commented on PR #18524: URL: https://github.com/apache/tvm/pull/18524#issuecomment-3590966049
## Summary of Changes Hello @Dayuxiaoshui, I'm Gemini Code Assist[^1]! 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 addresses a critical bug in the TVM frontend for PyTorch's `scaled_dot_product_attention` operation, specifically when processing 2D input tensors. Previously, the frontend would throw an `InternalError` due to an incorrect assumption of 4D input dimensions. The fix introduces logic to dynamically detect input dimensionality, reshaping 2D inputs to 4D for the attention computation and then restoring their original shape, while preserving the behavior for 4D inputs. This ensures broader compatibility and robustness for models utilizing `scaled_dot_product_attention` within the TVM framework, backed by a new test case and an adjustment to the model verification utility. ### Highlights * **Fix for 2D `scaled_dot_product_attention` inputs**: Resolved an `InternalError` that occurred when `scaled_dot_product_attention` received 2D inputs, as the TVM frontend incorrectly assumed 4D inputs. * **Dynamic input dimensionality handling**: The `_scaled_dot_product_attention` function now detects whether inputs are 2D or 4D. For 2D inputs, it expands them to 4D for processing and then squeezes them back to 2D. For 4D inputs, existing behavior is maintained. * **New test case for 2D inputs**: A dedicated test case (`Attention2D`) has been added to `test_frontend_from_exported_program.py` to verify the correct conversion and numerical stability of `scaled_dot_product_attention` with 2D inputs. * **`verify_model_numerically` update**: The `verify_model_numerically` helper function now uses `strict=False` when calling `torch.export` to accommodate operations like `scaled_dot_product_attention` that may have internal non-exportable components. <details> <summary><b>Using Gemini Code Assist</b></summary> <br> The full guide for Gemini Code Assist can be found on our [documentation page](https://developers.google.com/gemini-code-assist/docs/review-github-code), here are some quick tips. <b>Invoking Gemini</b> You can request assistance from Gemini at any point by creating a comment using either `/gemini <command>` or `@gemini-code-assist <command>`. Below is a summary of the supported commands on the current page. 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 pull request comments and review comments. Help | `/gemini help` | Displays a list of available commands. <b>Customization</b> 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](https://developers.google.com/gemini-code-assist/docs/customize-gemini-behavior-github). <b>Limitations & Feedback</b> Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with :thumbsup: and :thumbsdown: on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up [here](https://google.qualtrics.com/jfe/form/SV_2cyuGuTWsEw84yG). <b>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](https://cloud.google.com/products/gemini/code-assist).</b> </details> [^1]: Review the [Privacy Notices](https://policies.google.com/privacy), [Generative AI Prohibited Use Policy](https://policies.google.com/terms/generative-ai/use-policy), [Terms of Service](https://policies.google.com/terms), and learn how to configure Gemini Code Assist in GitHub [here](https://developers.google.com/gemini-code-assist/docs/customize-gemini-behavior-github). Gemini can make mistakes, so double check it and [use code with caution](https://support.google.com/legal/answer/13505487). -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
