app.py

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_community.vectorstores import FAISS
from langchain_google_genai import ChatGoogleGenerativeAI
from langchain.chains.question_answering import load_qa_chain
from langchain.prompts import PromptTemplate
from dotenv import load_dotenv

# Load environment variables
load_dotenv()

# Configure API Key
GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
if not GOOGLE_API_KEY:
    raise ValueError("API key not found in environment variables. Please check 'key.env'.")

genai.configure(api_key=GOOGLE_API_KEY)

# Helper Functions
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)
    return text_splitter.split_text(text)

def get_vector_store(text_chunks):
    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=GOOGLE_API_KEY)
    vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
    vector_store.save_local("faiss_index")

def get_conversational_chain():
    prompt_template = """
    {context}

    You are an AI Agri Farmer Advisor. Your role is to:
    1. Answer questions about farming, crop management, soil health, fertilizers, pest control, and more.
    2. Provide detailed instructions, tips, and advice based on the question.
    3. If information is not found in the document, answer based on your knowledge.

    ## Task:
    {question}

    Provide a detailed and helpful response.
    """
    model = ChatGoogleGenerativeAI(model="gemini-pro", temperature=0.3)
    prompt = PromptTemplate(template=prompt_template, input_variables=["context", "question"])
    chain = load_qa_chain(model, chain_type="stuff", prompt=prompt)
    return chain

def user_input(user_question):
    embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001", google_api_key=GOOGLE_API_KEY)
    
    faiss_index_path = "faiss_index/index.faiss"
    if not os.path.exists(faiss_index_path):
        return {"error": "FAISS index not found. Please ensure the documents are processed first."}
    
    new_db = FAISS.load_local("faiss_index", embeddings, allow_dangerous_deserialization=True)
    docs = new_db.similarity_search(user_question)

    chain = get_conversational_chain()
    response = chain({"input_documents": docs, "question": user_question}, return_only_outputs=True)

    return {"response": response["output_text"]}

# Streamlit UI
def main():
    st.set_page_config(page_title="🌾 AI Agri Farmer Chat Bot 🌿", page_icon="🌾")
    st.header("🌾 AI Agri Farmer Chat Bot 🌱")
    st.markdown("""
    Welcome to the AI Agri Farmer Chat Bot! 🤖 Ask any farming-related questions, get advice on crops, soil, fertilizers, pest management, and more.
    If your question is not covered in the document, the bot will answer using its knowledge base. 🌟
    """)
    
    # Input field for user's question
    user_query = st.text_input("💬 Ask your question about farming or agriculture:")
    st.caption("Example: 'What is the best fertilizer for cotton farming?'")

    if user_query:
        response = user_input(user_query)
        if "error" in response:
            st.error(response["error"])
        else:
            st.write("🤖 Reply: ", response["response"])

    # Sidebar for PDF processing
    with st.sidebar:
        st.title("📄 Document Information")
        st.info("The chatbot uses information from `ai-farming.pdf` to answer your questions.")

        if not os.path.exists("faiss_index/index.faiss"):
            with st.spinner("Processing the farming document..."):
                pdf_files = ["ai-farming.pdf"]
                raw_text = get_pdf_text([open(pdf, "rb") for pdf in pdf_files])
                text_chunks = get_text_chunks(raw_text)
                get_vector_store(text_chunks)
                st.success("Document processed successfully!")
        else:
            st.info("The document index is already loaded.")

if __name__ == "__main__":
    main()