RAG Integrated

app.py

@@ -1,68 +1,87 @@
 import streamlit as st
-from transformers import pipeline
-import fitz
+from langchain.document_loaders import PyPDFLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.vectorstores import Chroma
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.chains import RetrievalQA
+from langchain_google_genai import ChatGoogleGenerativeAI
+import tempfile
+import os
+from dotenv import load_dotenv
+from pydantic import SecretStr
 
-qa = pipeline("question-answering", model="deepset/roberta-base-squad2", device=0)
-text_gen = pipeline("text2text-generation", model="google/flan-t5-base", device=0)
 
+load_dotenv()
+GOOGLE_API_KEY = os.getenv("GOOGLE_API_KEY")
 
-# extract text from uploaded document
-def extract_PDF(file):
-    text = ""
-    with fitz.open(stream=file.read(), filetype="pdf") as doc:
-        for page in doc:
-            text += page.get_text()
-    return text
+embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
 
+# ---------------------------- SETUP ----------------------------
 
-# ------------------------------------------------------------------------------
+st.title("📄 LangChain RAG Chatbot")
 
-# -----------------------------------Streamlit UI--------------------------------
+# Session state
+if "chat_history" not in st.session_state:
+    st.session_state.chat_history = []
 
-st.title("Chatbot with Huggingface")
+if "qa_chain" not in st.session_state:
+    st.session_state.qa_chain = None
 
-st.subheader("Upload file")
+# ---------------------------- FILE UPLOAD ----------------------------
+
+st.subheader("Upload your PDF")
 pdf_file = st.file_uploader("Upload", type="pdf")
 
-# Initialize Session state for convo history
+if pdf_file is not None and st.session_state.qa_chain is None:
+    with st.spinner("🔍 Processing document..."):
+        # Save file temporarily
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".pdf") as tmp_file:
+            tmp_file.write(pdf_file.read())
+            tmp_path = tmp_file.name
 
-if "chat_history" not in st.session_state:
-    st.session_state.chat_history = []
+        # Load and split PDF
+        loader = PyPDFLoader(tmp_path)
+        documents = loader.load_and_split()
+
+        splitter = RecursiveCharacterTextSplitter(chunk_size=1000, chunk_overlap=200)
+        chunks = splitter.split_documents(documents)
 
-if "context" not in st.session_state:
-    st.session_state.context = None
+        # Vector store
+        
+        vectordb = Chroma.from_documents(
+            chunks, embeddings, persist_directory="./chroma_db"
+        )
+        retriever = vectordb.as_retriever()
 
-# extract text and store in the session
-if pdf_file is not None and st.session_state.context is None:
-    st.session_state.context = extract_PDF(pdf_file)
+        # QA Chain
+        llm = ChatGoogleGenerativeAI(model="gemini-2.0-flash", api_key=SecretStr(GOOGLE_API_KEY) if GOOGLE_API_KEY else None)
+        qa_chain = RetrievalQA.from_chain_type(llm=llm, retriever=retriever)
 
+        # Store in session
+        st.session_state.qa_chain = qa_chain
+        st.success("✅ Document processed and indexed!")
 
-# Chat section
+# ---------------------------- CHAT ----------------------------
 
-if st.session_state.context:
-    st.subheader("Chat with the PDF")
+if st.session_state.qa_chain:
+    st.subheader("💬 Ask a question")
 
-    question = st.text_input("You", key="user_input")
+    question = st.text_input("You:", key="user_input")
 
-    
     if question:
-        result = qa(question=question, context=st.session_state.context)
-        
-        context_chunk = st.session_state.context[:1500]
-        prompt = f"Context: {context_chunk}\nQuestion: {question}\nAnswer:"
-        
-        generated = text_gen(prompt, max_length=100)[0]['generated_text']
-        
-        # save convo
-        st.session_state.chat_history.append(
-            {"user": question, "bot": generated}
-        )
-
-        # Display chat
+        with st.spinner("🤖 Generating answer..."):
+            answer = st.session_state.qa_chain.run(question)
+            st.session_state.chat_history.append({"user": question, "bot": answer})
 
+    # Display chat history
     for chat in st.session_state.chat_history:
-        st.markdown(f"**You:** {chat['user']}")
-        st.markdown(f"**Bot:** {chat['bot']}")
-
+        st.markdown(f"🧑 **You:** {chat['user']}")
+        st.markdown(f"🤖 **Bot:** {chat['bot']}")
+
+    # Reset button
+    if st.button("🔄 Reset Chat"):
+        st.session_state.chat_history = []
+        st.session_state.qa_chain = None
+        st.rerun()
 else:
-    st.info("Please upload PDF to begin")
+    st.info("📂 Please upload a PDF to begin.")

app_bkp.py

@@ -1,37 +1,68 @@
 import streamlit as st
 from transformers import pipeline
-import fitz  # PyMuPDF
+import fitz
 
+qa = pipeline("question-answering", model="deepset/roberta-base-squad2", device=0)
+text_gen = pipeline("text2text-generation", model="google/flan-t5-base", device=0)
 
 
-question_answerer = pipeline(
-    "question-answering", model="distilbert-base-cased-distilled-squad"
-)
-
-# Extract text from uploaded PDF
-def extract_text_from_pdf(uploaded_file):
+# extract text from uploaded document
+def extract_PDF(file):
     text = ""
-    with fitz.open(stream=uploaded_file.read(), filetype="pdf") as doc:
+    with fitz.open(stream=file.read(), filetype="pdf") as doc:
         for page in doc:
-            text += page.get_text()
+            text += page.get_text() # type: ignore
     return text
 
 
-st.title("PDF Chatbot with Transformers")
+# ------------------------------------------------------------------------------
+
+# -----------------------------------Streamlit UI--------------------------------
+
+st.title("Chatbot with Huggingface")
+
+st.subheader("Upload file")
+pdf_file = st.file_uploader("Upload", type="pdf")
+
+# Initialize Session state for convo history
+
+if "chat_history" not in st.session_state:
+    st.session_state.chat_history = []
+
+if "context" not in st.session_state:
+    st.session_state.context = None
 
-st.subheader("Upload a PDF Document")
-pdf_file = st.file_uploader("Upload your PDF file", type="pdf")
+# extract text and store in the session
+if pdf_file is not None and st.session_state.context is None:
+    st.session_state.context = extract_PDF(pdf_file)
 
-if pdf_file is not None:
-    context = extract_text_from_pdf(pdf_file)
 
-    st.subheader("Ask a Question")
-    question = st.text_input("Your question:", "What is this document about?")
+# Chat section
 
+if st.session_state.context:
+    st.subheader("Chat with the PDF")
+
+    question = st.text_input("You", key="user_input")
+
+    
     if question:
-        result = question_answerer(question=question, context=context)
-        st.write("**Answer:**", result["answer"])
-else:
-    st.info("Please upload a PDF to begin.")
+        result = qa(question=question, context=st.session_state.context) # type: ignore
+        
+        context_chunk = st.session_state.context[:1500]
+        prompt = f"Context: {context_chunk}\nQuestion: {question}\nAnswer:"
+        
+        generated = text_gen(prompt, max_length=100)[0]['generated_text']  # type: ignore
 
+        # save convo
+        st.session_state.chat_history.append(
+            {"user": question, "bot": generated}
+        )
 
+        # Display chat
+
+    for chat in st.session_state.chat_history:
+        st.markdown(f"**You:** {chat['user']}")
+        st.markdown(f"**Bot:** {chat['bot']}")
+
+else:
+    st.info("Please upload PDF to begin")

requirements.txt

@@ -1,6 +1,12 @@
 streamlit
-torch
-accelerate
+openai
+langchain-google-genai
+langchain-core
+langchain-text-splitters
 transformers
 tf-keras
-PyMuPDF
+langchain 
+chromadb 
+tiktoken 
+pypdf
+sentence-transformers