store_index.py

from src.helper import load_pdf,text_split,download_hugging_face_embeddings
from langchain.vectorstores import Pinecone
import pinecone
from dotenv import load_dotenv
import os

load_dotenv()

PINECONE_API_KEY=os.environ.get('PINECONE_API_KEY')
PINECONE_API_ENV=os.environ.get('PINECONE_API_ENV')

#print(PINECONE_API_KEY)
#print(PINECONE_API_ENV)


#load the pdf
extracted_data=load_pdf("data/")

#apply text splitter
text_chunks=text_split(extracted_data)

#download the embeddings
embeddings=download_hugging_face_embeddings()

#Initializing the Pinecone
pinecone.init(api_key=PINECONE_API_KEY,
              environment=PINECONE_API_ENV)

#give index name
index_name="medical--chatbot"


#creating embeddings for each of the text chunks and storing
docsearch=Pinecone.from_texts([t.page_content for t in text_chunks],  embeddings,index_name=index_name)