Will Fu-Hinthorn

Thanks for the PR! The reason we didn't use sklearn is we don't want to add it as a required dependency. We could add an import check at the top...

Thanks for the PR! The reason we didn't use sklearn is we don't want to add it as a required dependency. We could add an import check at the top...

gpt-3.5-turbo is optimized for chat rather than schema generation, so it is less reliable. Recommend trying out the openai functions agent or use a different model.

LGTM! I'll give it another run before merging

For "await" - it has to be done within an outer async function `async def main(): ....`

Right! In notebooks there is already an event loop running, but in the exported python script that breaks down, so you'd have to either wrap in an async call or...

What are your python and playwright and pydantic versions

Thanks for flagging this @YYYasin19 , and thank you for your patience! I'll work on resolving this. We moved some callbacks schemas to a shared package to let people use...

Opened a PR to add to conda https://github.com/conda-forge/staged-recipes/pull/23151 We'll publish the SDK repo. It's used for tracing schemas that are shared with other providers (e.g., WandB) so I'm not going...

Made the MIT-licensed code public https://github.com/langchain-ai/langchainplus-sdk Timeline for incorporation in conda is now dependent on their review cycles.