Harrison Chase

Results 367 comments of Harrison Chase

using nearText or nearVector to request weaviate

> > hey @ZouhairElhadi, thanks for contribution! what's the use case you're imagining? why have an attribute on the VectorStore that determines which similarity search to use versus just letting...

InvalidRequestError: logprobs, best_of and echo parameters are no…

@jcomfort4 what embedding model are you using? i think the issue is the embedding model has a different api base than the chat model and the enviroment is getting confused

InvalidRequestError: logprobs, best_of and echo parameters are no…

> I guess in the simplified version I'm not using (or not specifying) an embedding API, but maybe I need to be? For example, following a sample like this from...

Add a new search_type option `similarity_with_score` to VectorStore

added in other pr!

Feature/4257 enum output parser

adding https://github.com/hwchase17/langchain/pull/5165 in place of this

Do not overwrite openai.api_version.

@zijie0 sorry for delay! how does https://github.com/hwchase17/langchain/pull/4687 look?

fix for azure open ai not returning content sometimes for the when stopping.

hmmm do we know why the response is missing content?

Fix #4087 by setting the correct csv dialect

> Lol yup. Maybe I did over engineer this. I'm traveling right now, will push changes in the evening. Till then, any comment on my 2nd open question? I feel...

Fix #4087 by setting the correct csv dialect

@rick2047 sorry for delay - looks great~!

Add Chain equivalents of Map() and Reduce()

@IndikaUdagedara all chains have an `apply` method on them - that covers the `map` use case right? could you motivate the `reduce` use case a bit more?