Added Pinecone serverless with delta conversations

SartajBhuvaji / Resonate

Data Science Capstone Project based on RAG LLMs. The project aims to improve meetings by providing an interface to recollect information from audio/video meetings.

https://huggingface.co/spaces/sartajbhuvaji/Resonate

MIT License

6 stars 1 forks source link

Added Pinecone serverless with delta conversations #22

Closed SartajBhuvaji closed 6 months ago

SartajBhuvaji commented 7 months ago

Added src/db/pinecone_serverless.py

Using Pinecone serverless to upload data, query it along with the delta window Also added a parser that parses the delta conversations into a data frame.

madhuroopa commented 7 months ago

Hey @SartajBhuvaji,

As we need to store the corresponding video file of a meeting in the firebase, there should be another key value pair in the JSON document for the video file storage. Can you please add that to the JSON doc structure.

SartajBhuvaji commented 7 months ago

Hey,

Sure!

I'll update the JSON to also include a boolean field for each meeting to indicate if the meeting has a video file linked
I'll add a function to search all namespaces
And a function to get all the complete transcripts. This should be helpful in case of summarizing the meetings.

SartajBhuvaji commented 7 months ago

@madhuroopa Fixed Issue #20 Missing pinecone namespace for the final upstreams

Also, updated the JSON store object. (attached demo JSON store file) namespace2.json

SartajBhuvaji commented 7 months ago

@PrachiteeChouhan31 Added the query_every_namespace(query) function that queries all namespaces in the index.

madhuroopa commented 7 months ago

Hey @SartajBhuvaji,

Can you add small docstrings to each method for quicker understanding.

SartajBhuvaji commented 7 months ago

@madhuroopa

Added comments for important functions.