app.py

import os
import streamlit as st
import fitz  # PyMuPDF
from sentence_transformers import SentenceTransformer
from pinecone import Pinecone
from langchain_community.embeddings.sentence_transformer import SentenceTransformerEmbeddings
from langchain_pinecone import PineconeVectorStore
from langchain_text_splitters import CharacterTextSplitter

# Load the sentence transformer model
model = SentenceTransformer('all-MiniLM-L6-v2')

def pdf_to_text(pdf_file):
    doc = fitz.open(stream=pdf_file.read(), filetype="pdf")
    text = ""
    for page in doc:
        text += page.get_text()
    return text

st.title("PDF to Embeddings")

# User inputs for Pinecone
api_key = st.text_input("Enter your Pinecone API Key", type="password")
index_name = st.text_input("Enter the Pinecone Index Name")

# File uploader
uploaded_file = st.file_uploader("Upload a PDF file", type="pdf")

if uploaded_file is not None:
    # Extract text from PDF
    text = pdf_to_text(uploaded_file)
    st.write("Extracted Text:")
    st.write(text)
    
    text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=0)
    docs = text_splitter.split_text(text)

    # Initialize Pinecone Index
    if api_key and index_name:
        pc = Pinecone(
            api_key=api_key
        )
        index = pc.Index(index_name)
        embeddings = SentenceTransformerEmbeddings(model_name="all-MiniLM-L6-v2")
        
        os.environ['PINECONE_API_KEY'] = api_key
        
        docsearch = PineconeVectorStore.from_texts(docs, embeddings, index_name=index_name)

        st.write(f"Successfully uploaded {len(docs)} vectors to Pinecone Index '{index_name}'")
    else:
        st.write("Please provide your Pinecone API Key and Index Name.")