Devin Stein comments

Results 94 comments of


                                            Devin Stein

TypeError: issubclass() arg 1 must be a class

@getsean Look at your account page: https://platform.openai.com/account/usage

SQLDatabase and SQLDatabaseChain with AWS Athena

@clebermarq What version of SQLAlchemy are you using? It looks like [PyAthena only supports

map_reduce and refine not working with RetrievalQA chain

@KillerStrike17 Can you share error logs?

Llama_index model as a tool for lang chain

@vladimircape The bot's answer is correct about the error message. You cannot pass the LlamaIndex model to `initialize_agent` ```py agent_executor = initialize_agent( # llm must be a LangChain LLM tools,...

Llama_index model as a tool for lang chain

@vladimircape Can you share code?

Issue: Can not import the Langchain modules.

@shivam-36 The latest version of LangChain is `0.0.222`. Try upgrade to the latest and see if that works.

ValidationError: 1 validation error for ConversationalRetrievalChain chain_type_kwargs extra fields not permitted (type=value_error.extra)

_Answer generated by a 🤖_ ### Answer The error you're encountering is due to the fact that you're passing the `chain_type_kwargs` dictionary directly to the `ConversationalRetrievalChain.from_llm()` method. Instead, you should...

Can not import NotRequired from typing_extensions.py

_Generated by a 🤖_ ## Answer I found related issues that might be helpful. I did my best to summarize the solutions, but I recommend looking at the related issues...

why llm_cache only supports in BaseLLM, not in BaseChatModel? I can't use llm_cache when using ChatOpenAI. instead, i can only use it by using OpenAI.

_Generated by a 🤖_ ### Answer The `llm_cache` is not supported in `BaseChatModel` because it is designed specifically for the BaseLLM class, which focuses on generating text completions with a...

ValueError: invalid literal for int() with base 10: '0<|im_end|>' use it will throw an exception

_Generated by a 🤖_ ### Answer The issue seems to be related to the parsing of the output from the language model. The output contains a special token ' ###...