Spaces:
Running
Running
File size: 3,325 Bytes
45e98b3 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 |
import streamlit as st
from PyPDF2 import PdfReader
from langchain.text_splitter import RecursiveCharacterTextSplitter
import os
from langchain_google_genai import GoogleGenerativeAIEmbeddings
import google.generativeai as genai
from langchain.vectorstores import Pinecone as PC
from langchain_google_genai import ChatGoogleGenerativeAI
from dotenv import load_dotenv
load_dotenv()
os.getenv("GOOGLE_API_KEY")
os.getenv("PINECONE_API_KEY")
genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
os.environ['PINECONE_API_KEY'] = os.getenv("PINECONE_API_KEY")
def Pine():
from pinecone import Pinecone, ServerlessSpec
pc = Pinecone(api_key=os.getenv("PINECONE_API_KEY"))
index_name = "testing"
if index_name not in pc.list_indexes().names():
pc.create_index(
name=index_name,
dimension=768,
metric="cosine",
spec=ServerlessSpec(cloud='aws', region='us-east-1')
)
return index_name
def get_pdf_text(pdf_docs):
text = ""
for pdf in pdf_docs:
pdf_reader = PdfReader(pdf)
for page in pdf_reader.pages:
text += page.extract_text()
return text
def get_text_chunks(text):
text_splitter = RecursiveCharacterTextSplitter(chunk_size=10000, chunk_overlap=1000)
chunks = text_splitter.split_text(text)
return chunks
def get_vector_store(text_chunks):
index_name = Pine()
embedding = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
docsearch = PC.from_texts([t for t in text_chunks], embedding, index_name=index_name)
return docsearch
def showman(pdf_docs):
st.header("Chat with PDF")
user_question = st.text_input("Ask a Question from the PDF Files", key="user_question")
ask_another_question = st.button("Ask Another Question",on_click=clear_text)
if user_question and not ask_another_question:
llm = ChatGoogleGenerativeAI(model="models/gemini-1.5-pro-latest", temperature=0.9)
from langchain.chains import RetrievalQA
qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=st.session_state["docsearch"].as_retriever())
response = qa(user_question)
st.session_state["response"] = response["result"]
st.write("Answer:", st.session_state["response"])
def clear_text():
st.session_state["user_question"] = ""
st.session_state["response"] = ""
def show():
with st.sidebar:
st.title("Menu:")
pdf_docs = st.file_uploader("Upload your PDF Files", accept_multiple_files=True)
st.session_state["pdf_docs"] = pdf_docs if pdf_docs is not None else st.session_state.get("pdf_docs", [])
processed = st.session_state.get("processed", False)
if not processed and pdf_docs:
if st.button("Submit & Process"):
with st.spinner("Processing..."):
raw_text = get_pdf_text(pdf_docs)
text_chunks = get_text_chunks(raw_text)
docsearch = get_vector_store(text_chunks)
st.session_state["docsearch"] = docsearch
st.session_state["processed"] = True
st.success("Done!")
showman(st.session_state["pdf_docs"]) |