I wonder if I am doing something

aashishsha

I wonder if I am doing something incorrect -- but all I did is added bunch of metadata properties

15 comments

LLogan M

Ohhh you have an outdated version of the code

LLogan M

this is fixed in another version

LLogan M

The new code does

Plain Text

        for i, entry in enumerate(entries):
            if i < query.similarity_top_k:
                similarities.append(get_node_similarity(entry, similarity_key))
                nodes.append(to_node(entry, text_key=self.text_key))
                node_idxs.append(str(i))
            else:
                break

LLogan M

So similarities are first

aashishsha

I am on 0.9.8

aashishsha

which version should I use

LLogan M

Hmm that fix didn't end up making it into the changelog. I would just use the latest if you are already on v0.9.8

aashishsha

cool thanks a ton

aashishsha

I was looking at changelog

aashishsha

@Logan M are there any breaking changes in 0.9.12 version for open AI

aashishsha

File ~/opt/anaconda3/envs/LLMTools/lib/python3.9/site-packages/llama_index/llms/openai.py:103, in OpenAI()
95 timeout: float = Field(
96 default=60.0,
97 description="The timeout, in seconds, for API requests.",
98 gte=0,
99 )
100 default_headers: Dict[str, str] = Field(
101 default=None, description="The default headers for API requests."
102 )
--> 103 reuse_client: bool = Field(
104 default=True,
105 description=(
106 "Reuse the OpenAI client between requests. When doing anything with large "
107 "volumes of async API calls, setting this to false can improve stability.",
108 ),
109 )
111 api_key: str = Field(default=None, description="The OpenAI API key.", exclude=True)
112 api_base: str = Field(description="The base URL for OpenAI API.")

TypeError: Argument 'description' has incorrect type (expected str, got tuple)

LLogan M

Yea, v0.9.13 is already out and fixes that I think right? Or did we not put 0.9.13 on pypi yet?

aashishsha

never mind 0.9.13 fixed it

LLogan M

Ah perfect

aashishsha

few challanges of being on cutting edge 🙂

Add a reply

Find answers from the community

I wonder if I am doing something