[smg][ci] migrate function calling tests to new infrastructure#16748
[smg][ci] migrate function calling tests to new infrastructure#16748
Conversation
Summary of ChangesHello @slin1237, 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 significantly refactors the end-to-end testing framework for function calling and tool choice features. By moving these tests to a new Highlights
🧠 New Feature in Public Preview: You can now enable Memory to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console. Using Gemini Code AssistThe 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 by creating a comment using either
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 Limitations & Feedback Gemini Code Assist 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. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here. 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
|
There was a problem hiding this comment.
Code Review
This pull request successfully migrates the function calling and tool choice end-to-end tests to a new pytest-based infrastructure, which is a great improvement for test maintainability and clarity. The new tests are comprehensive and cover a wide range of scenarios across different models and parsers. I've identified one logic bug in a test case that could lead to false positives and another area where the code for reconstructing streaming responses could be made more robust. I've provided specific suggestions for both points. Overall, this is a solid contribution to improving the testing framework.
a564965 to
e248fe3
Compare
Migrate function calling and tool choice tests from e2e_grpc to chat_completions using the new pytest fixture infrastructure. Tests migrated from test_openai_function_calling.py: - TestOpenAIServerFunctionCalling (llama-1b, llama parser): 10 tests for function calling format, streaming, and tool_choice options - TestOpenAIPythonicFunctionCalling (llama-8b, pythonic parser): 2 tests for pythonic function calling format Tests migrated from test_tool_choice.py: - TestToolChoiceLlama (llama-1b, llama parser): 13 tests for tool_choice auto/required/specific modes with streaming and non-streaming - TestToolChoiceQwen (qwen-7b, qwen parser): inherits all tests - TestToolChoiceMistral (mistral-7b, mistral parser): inherits all tests with test_complex_parameters_required_non_streaming skipped Total: 25 test methods across 5 test classes using 4 different models.
Migrate function calling and tool choice tests from e2e_grpc to chat_completions using the new pytest fixture infrastructure.
Tests migrated from test_openai_function_calling.py:
Tests migrated from test_tool_choice.py:
Total: 25 test methods across 5 test classes using 4 different models.
Checklist
Review Process
/tag-run-ci-label,/rerun-failed-ci,/tag-and-rerun-ci) or contact authorized users to do so.