app.py

!pip install streamlit langchain chromadb unstructured faiss-cpu sentence_transformers PyPDF2 groq
!pip install -U langchain-community
import os
os.environ["GROQ_API_KEY"] = "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"  # Replace with your key
import streamlit as st
from langchain.document_loaders import PyPDFLoader
from langchain.text_splitter import RecursiveCharacterTextSplitter
from langchain.embeddings import SentenceTransformerEmbeddings
from langchain.vectorstores import FAISS
import os
from groq import Groq

# Load PDF (with error handling)
def load_pdf(uploaded_file):
    try:
        loader = PyPDFLoader(uploaded_file)
        documents = loader.load()
        return documents
    except Exception as e:
        st.error(f"Error loading PDF: {e}")
        return None

# Chunking (with error handling)
def chunk_text(documents):
    try:
        text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
        chunks = text_splitter.split_documents(documents)
        return chunks
    except Exception as e:
        st.error(f"Error chunking text: {e}")
        return None

# Embeddings and Vectorstore (with error handling)
def create_embeddings_and_store(chunks):
    try:
        embeddings = SentenceTransformerEmbeddings(model_name="all-mpnet-base-v2")
        db = FAISS.from_documents(chunks, embeddings)
        return db
    except Exception as e:
        st.error(f"Error creating embeddings: {e}")
        return None

# Groq interaction (with more robust error handling)
def query_groq(query, db):
    try:
        docs = db.similarity_search(query)
        context = "\n".join([doc.page_content for doc in docs])

        client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
        if not client.api_key:  # Check if API key is set
            st.error("GROQ_API_KEY environment variable is not set.")
            return None

        prompt = f"""Use the following context to answer the question: {query}\n\nContext:\n{context}"""

        chat_completion = client.chat.completions.create(
            messages=[{"role": "user", "content": prompt}],
            model="llama-3.3-70b-versatile",  # Or other suitable open-source model
        )
        return chat_completion.choices[0].message.content
    except Exception as e:
        st.error(f"Error querying Groq: {e}")
        return None


# Streamlit app
st.title("RAG Application")

uploaded_file = st.file_uploader("Upload PDF", type="pdf")

if uploaded_file is not None:
    with st.spinner("Processing PDF..."):
        documents = load_pdf(uploaded_file)
        if documents: # Check if PDF loaded successfully
            chunks = chunk_text(documents)
            if chunks: # Check if chunks were created successfully
                db = create_embeddings_and_store(chunks)
                if db: # Check if embeddings were created successfully
                    st.success("PDF processed!")

                    query = st.text_area("Enter your query")
                    if st.button("Submit"):
                        if query:
                            with st.spinner("Querying..."):
                                answer = query_groq(query, db)
                                if answer: # Check if query was successful
                                    st.write(answer)
                        else:
                            st.warning("Please enter a query.")