Update app.py

app.py

@@ -1,246 +1,123 @@
-# import os
-# import logging
-# from dotenv import load_dotenv
-# import streamlit as st
-# from PyPDF2 import PdfReader
-# from langchain.text_splitter import CharacterTextSplitter
-# # from langchain.embeddings import HuggingFaceInstructEmbeddings
-# from langchain_cohere import CohereEmbeddings
-# from langchain.vectorstores import FAISS
-# from langchain.memory import ConversationBufferMemory
-# from langchain.chains import ConversationalRetrievalChain
-# # from langchain.llms import Ollama
-# from langchain_groq import ChatGroq
-
-# # Load environment variables
-# load_dotenv()
-
-# # Set up logging
-# logging.basicConfig(
-#     level=logging.INFO,
-#     format='%(asctime)s - %(levelname)s - %(message)s'
-# )
-
-# # Function to extract text from PDF files
-# 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
-
-# # Function to split the extracted text into chunks
-# def get_text_chunks(text):
-#     text_splitter = CharacterTextSplitter(
-#         separator="\n",
-#         chunk_size=1000,
-#         chunk_overlap=200,
-#         length_function=len
-#     )
-#     chunks = text_splitter.split_text(text)
-#     return chunks
-
-# # Function to create a FAISS vectorstore
-# # def get_vectorstore(text_chunks):
-# #     embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl")
-# #     vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
-# #     return vectorstore
-
-# def get_vectorstore(text_chunks):
-#     cohere_api_key = os.getenv("COHERE_API_KEY")
-#     embeddings = CohereEmbeddings(model="embed-english-v3.0", cohere_api_key=cohere_api_key)
-#     vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
-#     return vectorstore
-
-# # Function to set up the conversational retrieval chain
-# def get_conversation_chain(vectorstore):
-#     try:
-#         # llm = Ollama(model="llama3.2:1b")
-#         llm = ChatGroq(model="llama-3.3-70b-versatile", temperature=0.5)
-#         memory = ConversationBufferMemory(memory_key='chat_history', return_messages=True)
-        
-#         conversation_chain = ConversationalRetrievalChain.from_llm(
-#             llm=llm,
-#             retriever=vectorstore.as_retriever(),
-#             memory=memory
-#         )
-        
-#         logging.info("Conversation chain created successfully.")
-#         return conversation_chain
-#     except Exception as e:
-#         logging.error(f"Error creating conversation chain: {e}")
-#         st.error("An error occurred while setting up the conversation chain.")
-
-# # Handle user input
-# def handle_userinput(user_question):
-#     if st.session_state.conversation is not None:
-#         response = st.session_state.conversation({'question': user_question})
-#         st.session_state.chat_history = response['chat_history']
-
-#         for i, message in enumerate(st.session_state.chat_history):
-#             if i % 2 == 0:
-#                 st.write(f"*User:* {message.content}")
-#             else:
-#                 st.write(f"*Bot:* {message.content}")
-#     else:
-#         st.warning("Please process the documents first.")
-
-# # Main function to run the Streamlit app
-# def main():
-#     load_dotenv()
-#     st.set_page_config(page_title="Chat with multiple PDFs", page_icon=":books:")
-
-#     if "conversation" not in st.session_state:
-#         st.session_state.conversation = None
-#     if "chat_history" not in st.session_state:
-#         st.session_state.chat_history = None
-
-#     st.header("Chat with multiple PDFs :books:")
-#     user_question = st.text_input("Ask a question about your documents:")
-#     if user_question:
-#         handle_userinput(user_question)
-
-#     with st.sidebar:
-#         st.subheader("Your documents")
-#         pdf_docs = st.file_uploader(
-#             "Upload your PDFs here and click on 'Process'", accept_multiple_files=True
-#         )
-#         if st.button("Process"):
-#             with st.spinner("Processing..."):
-#                 raw_text = get_pdf_text(pdf_docs)
-#                 text_chunks = get_text_chunks(raw_text)
-#                 vectorstore = get_vectorstore(text_chunks)
-#                 st.session_state.conversation = get_conversation_chain(vectorstore)
-
-# if __name__ == '__main__':
-#     main()
-
-
-
-
-
-
-
-import streamlit as st
 import os
+import logging
 from dotenv import load_dotenv
-import PyPDF2
-import requests
-import cohere
-from langchain_text_splitters import RecursiveCharacterTextSplitter
-from langchain_community.vectorstores import FAISS
+import streamlit as st
+from PyPDF2 import PdfReader
+from langchain.text_splitter import CharacterTextSplitter
+# from langchain.embeddings import HuggingFaceInstructEmbeddings
 from langchain_cohere import CohereEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.memory import ConversationBufferMemory
+from langchain.chains import ConversationalRetrievalChain
+# from langchain.llms import Ollama
+from langchain_groq import ChatGroq
 
 # Load environment variables
 load_dotenv()
-GEMINI_API_KEY = os.getenv("GEMINI_API_KEY")
-COHERE_API_KEY = os.getenv("COHERE_API_KEY")
 
-# Initialize Cohere client
-co = cohere.Client(COHERE_API_KEY)
+# Set up logging
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(levelname)s - %(message)s'
+)
 
-# Configure Streamlit
-st.set_page_config(page_title="RAG Chatbot with Gemini & Cohere")
-st.title("🤖 Multi-Model RAG Chatbot")
-
-# Initialize session state
-if "messages" not in st.session_state:
-    st.session_state.messages = []
-if "vector_store" not in st.session_state:
-    st.session_state.vector_store = None
-
-# File upload and processing
-uploaded_file = st.file_uploader("Upload a PDF document", type="pdf")
-
-if uploaded_file and not st.session_state.vector_store:
-    # Process PDF
-    pdf_reader = PyPDF2.PdfReader(uploaded_file)
+# Function to extract text from PDF files
+def get_pdf_text(pdf_docs):
     text = ""
-    for page in pdf_reader.pages:
-        text += page.extract_text()
-
-    # Split text
-    text_splitter = RecursiveCharacterTextSplitter(
+    for pdf in pdf_docs:
+        pdf_reader = PdfReader(pdf)
+        for page in pdf_reader.pages:
+            text += page.extract_text()
+    return text
+
+# Function to split the extracted text into chunks
+def get_text_chunks(text):
+    text_splitter = CharacterTextSplitter(
+        separator="\n",
         chunk_size=1000,
-        chunk_overlap=200
+        chunk_overlap=200,
+        length_function=len
     )
     chunks = text_splitter.split_text(text)
+    return chunks
 
-    # Create embeddings and vector store
-    embeddings = CohereEmbeddings(
-        cohere_api_key=COHERE_API_KEY,
-        model="embed-english-v3.0",
-        user_agent="rag-chatbot-v1"
-    )
-    st.session_state.vector_store = FAISS.from_texts(
-        texts=chunks,
-        embedding=embeddings
-    )
-
-# Display chat messages
-for message in st.session_state.messages:
-    with st.chat_message(message["role"]):
-        st.markdown(message["content"])
+# Function to create a FAISS vectorstore
+# def get_vectorstore(text_chunks):
+#     embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl")
+#     vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
+#     return vectorstore
 
-# Query expansion function
-def expand_query(query):
-    prompt = f"""Generate 3 query variations that help answer: {query}
-    Format as numbered bullet points:"""
+def get_vectorstore(text_chunks):
+    cohere_api_key = os.getenv("COHERE_API_KEY")
+    embeddings = CohereEmbeddings(model="embed-english-v3.0", cohere_api_key=cohere_api_key)
+    vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
+    return vectorstore
+
+# Function to set up the conversational retrieval chain
+def get_conversation_chain(vectorstore):
+    try:
+        # llm = Ollama(model="llama3.2:1b")
+        llm = ChatGroq(model="llama-3.3-70b-versatile", temperature=0.5)
+        memory = ConversationBufferMemory(memory_key='chat_history', return_messages=True)
+        
+        conversation_chain = ConversationalRetrievalChain.from_llm(
+            llm=llm,
+            retriever=vectorstore.as_retriever(),
+            memory=memory
+        )
+        
+        logging.info("Conversation chain created successfully.")
+        return conversation_chain
+    except Exception as e:
+        logging.error(f"Error creating conversation chain: {e}")
+        st.error("An error occurred while setting up the conversation chain.")
 
-    response = co.generate(
-        prompt=prompt,
-        max_tokens=100,
-        temperature=0.7
-    )
-    expanded_queries = [query] + [q.split(". ")[1] for q in response.generations[0].text.split("\n") if q]
-    return expanded_queries
+# Handle user input
+def handle_userinput(user_question):
+    if st.session_state.conversation is not None:
+        response = st.session_state.conversation({'question': user_question})
+        st.session_state.chat_history = response['chat_history']
 
-# Gemini API call
-def generate_with_gemini(context, query):
-    url = f"https://generativelanguage.googleapis.com/v1beta/models/gemini-2.0-flash:generateContent?key={GEMINI_API_KEY}"
+        for i, message in enumerate(st.session_state.chat_history):
+            if i % 2 == 0:
+                st.write(f"*User:* {message.content}")
+            else:
+                st.write(f"*Bot:* {message.content}")
+    else:
+        st.warning("Please process the documents first.")
 
-    system_prompt = f"""You're an expert assistant. Use this context to answer:
-    {context}
+# Main function to run the Streamlit app
+def main():
+    load_dotenv()
+    st.set_page_config(page_title="Chat with multiple PDFs", page_icon=":books:")
 
-    Apply Chain of Abstraction and Grounding (CAG):
-    1. Identify key concepts
-    2. Create abstract relationships
-    3. Ground in specific examples
-    4. Synthesize final answer"""
+    if "conversation" not in st.session_state:
+        st.session_state.conversation = None
+    if "chat_history" not in st.session_state:
+        st.session_state.chat_history = None
 
-    headers = {"Content-Type": "application/json"}
-    data = {
-        "contents": [{
-            "parts": [{
-                "text": f"{system_prompt}\n\nQuestion: {query}"
-            }]
-        }]
-    }
+    st.header("Chat with multiple PDFs :books:")
+    user_question = st.text_input("Ask a question about your documents:")
+    if user_question:
+        handle_userinput(user_question)
 
-    response = requests.post(url, json=data, headers=headers)
-    return response.json()["candidates"][0]["content"]["parts"][0]["text"]
+    with st.sidebar:
+        st.subheader("Your documents")
+        pdf_docs = st.file_uploader(
+            "Upload your PDFs here and click on 'Process'", accept_multiple_files=True
+        )
+        if st.button("Process"):
+            with st.spinner("Processing..."):
+                raw_text = get_pdf_text(pdf_docs)
+                text_chunks = get_text_chunks(raw_text)
+                vectorstore = get_vectorstore(text_chunks)
+                st.session_state.conversation = get_conversation_chain(vectorstore)
 
-# Chat input
-if prompt := st.chat_input("Ask about the document"):
-    st.session_state.messages.append({"role": "user", "content": prompt})
+if __name__ == '__main__':
+    main()
 
-    with st.chat_message("user"):
-        st.markdown(prompt)
 
-    # Query expansion
-    expanded_queries = expand_query(prompt)
 
-    # Retrieve documents
-    docs = []
-    for query in expanded_queries:
-        docs.extend(st.session_state.vector_store.similarity_search(query, k=2))
 
-    # Generate response
-    context = "\n\n".join([doc.page_content for doc in docs])
-    response = generate_with_gemini(context, prompt)
 
-    with st.chat_message("assistant"):
-        st.markdown(response)
 
-    st.session_state.messages.append({"role": "assistant", "content": response})