svpino / youtube-rag

Building a RAG application from scratch

This is a step-by-step guide to building a simple RAG (Retrieval-Augmented Generation) application using Pinecone and OpenAI's API. The application will allow you to ask questions about any YouTube video.

Setup

1. Create a virtual environment and install the required packages:

$ python3 -m venv .venv
$ source .venv/bin/activate
$ pip install -r requirements.txt

2. Create a free Pinecone account and get your API key from here.

3. Create a .env file with the following variables:

OPENAI_API_KEY = [ENTER YOUR OPENAI API KEY HERE]
PINECONE_API_KEY = [ENTER YOUR PINECONE API KEY HERE]
PINECONE_API_ENV = [ENTER YOUR PINECONE API ENVIRONMENT HERE]

youtube-rag's Issues

No active indexes found in your Pinecone project

I set up Pinecone, added both the API Key and the environment variables but seem to have issues with connecting to pinecone.

Any thoughts about how to modify this block of code or troubleshoot?

I do see the Jupyter variables for documents, embeddings and index_name in VS Code but not sure if Pinecone is being set correctly

from langchain_pinecone import PineconeVectorStore

index_name = "youtube-index"

pinecone = PineconeVectorStore.from_documents(
documents, embeddings, index_name=index_name
)

ERror:

390 f"Index '{index_name}' not found in your Pinecone project. "
391 f"Did you mean one of the following indexes: {', '.join(index_names)}"
392 )

ValueError: No active indexes found in your Pinecone project, are you sure you're using the right Pinecone API key and Environment? Please double check your Pinecone dashboard.

UnauthorizedException issue when Pinecone.from_documents()

Hi I found the pinecode connection has unauthorized issue, took some time and got solutions, hope it can help other learners.

1. If you use free tier in Pinecone, the ENV set like this:
   PINECONE_API_ENV="us-central1-gcp"
2. from langchain_pinecone import PineconeVectorStore caused UnauthorizedException issue, when I changed the following, it works well.

from langchain_pinecone import Pinecone

import os 
os.environ['PINECONE_API_KEY'] = "PINECONE_API_KEY"
index_name = "youtube-index"

pinecone = Pinecone.from_documents(         index_name = index_name, 
                                            documents = documents, 
                                            embedding = embeddings)

Thank you for this tutorial.

AttributeError: module 'whisper' has no attribute 'load_model'

I have installed all the requirements but I keep getting this error in the whisper function when I want to get the transcription
I'm using python 3.9..6

how to add openai's function calling also

in the example rag.py, how can we add openai's function calling also for external api.