Why SelfQueryRetriever.fromLLM is using await?

[mlegy]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

const selfQueryRetriever = await SelfQueryRetriever.fromLLM({
   llm: new ChatOpenAI(),
   vectorStore: new PineconeStore(),
   documentContents: "Brief summary of a movie",
   attributeInfo: [],
   structuredQueryTranslator: new PineconeTranslator(),
 });

Description

The documentation and tests for SelfQueryRetriever.fromLLM are always using await. I can not find the reason as it's synchronous method—it directly returns a new instance of SelfQueryRetriever without involving any promises that would necessitate the use of await.

Is that a mistake or this is intended? If it's a mistake, I will happily raise a PR to fix it.

Thanks!

System Info

Version: 0.1.33

[dosubot[bot]]

The use of await with SelfQueryRetriever.fromLLM in the documentation is indeed unnecessary since the method is synchronous and does not return a Promise. It should be called without await as follows:

const selfQueryRetriever = SelfQueryRetriever.fromLLM({
   llm: new ChatOpenAI(),
   vectorStore: new PineconeStore(),
   documentContents: "Brief summary of a movie",
   attributeInfo: [],
   structuredQueryTranslator: new PineconeTranslator(),
});

This oversight in the documentation could lead to confusion, and your initiative to correct it by proposing a pull request is valuable. Since the method directly returns a new instance of SelfQueryRetriever without involving any asynchronous operations, removing await from the examples and tests will clarify its correct usage.

Sources

• langchain/src/retrievers/self_query/index.ts

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[mlegy]

Thank you for confirming! Raised a PR here #5097