code files for Talk with multiple PDF Documents
🕑 Added 2023-04-29 12:00:39 +0000 UTCUPDATE: Fixed the error in streamlit version for combined index selection (thanks to Shubham184 from discord!) Also included all requirements now.
PAY attention to the size of your pdf files as it will cost tokens to embed them!!!
you might want to implement a button to the querry as everytime the script refreshes currently it will rerun the script.
I believe by default the current implementation is using text-davinci-003 as it is not the chat endpoint. You might want to set the temperature to 0 For the model.
search all echohive videos and patreon code: https://www.echohive.live/
this is for video: https://youtu.be/qiRSx7RnGjA
NOTE: this code is a bit rough around the edges but here it is anyways. Expect some errors to debug.
Comments
Greg Meadows
File "/Users/greg/Downloads/downloads-2TB/code-repos/app-pdf/modify.py", line 157, in OpenAIEmbeddings @root_validator() ^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/site-packages/pydantic/deprecated/class_validators.py", line 228, in root_validator raise PydanticUserError( pydantic.errors.PydanticUserError: If you use `@root_validator` with pre=False (the default) you MUST specify `skip_on_failure=True`. Note that `@root_validator` is deprecated and should be replaced with `@model_validator`. For further information visit https://errors.pydantic.dev/2.4/u/root-validator-pre-skip
Echo Hive
This usually means “langchain” is not pip installed in the environment you are running the python file from.
Greg Meadows
ModuleNotFoundError: No module named 'langchain'