add source code

.gitignore

@@ -0,0 +1,2 @@
+*venv
+/faiss_index

app.py

@@ -0,0 +1,62 @@
+import streamlit as st
+from src.pdf_processing import extract_pdf_text, split_text_into_chunks
+from src.vector_store import create_and_save_vector_store
+from src.query_handler import handle_user_query
+
+# Initialize session state for chat history
+def initialize_session_state():
+    if 'messages' not in st.session_state:
+        st.session_state.messages = []
+
+def main():
+    """
+    Main function to run the Streamlit app.
+    """
+    initialize_session_state()
+    
+    st.set_page_config("DocuChat")
+    st.header("DocuChat: Chat with your Document")
+    st.markdown("Source code available at [[GitHub]](https://github.com/TejaCherukuri/DocuChat)")
+    
+    # Display previous chat messages
+    for message in st.session_state.messages:
+        with st.chat_message(message["role"]):
+            st.write(message["content"])
+
+    # Chat input for user questions
+    if prompt := st.chat_input("Ask a question about your document"):
+        st.session_state.messages.append({"role": "user", "content": prompt})
+        
+        with st.chat_message("user"):
+            st.write(prompt)
+
+        with st.chat_message("assistant"):
+            with st.spinner("Thinking..."):
+                try:
+                    response = handle_user_query(prompt)
+                    st.write(response)
+                    
+                    # Save assistant's response
+                    st.session_state.messages.append({"role": "assistant", "content": response})
+                except Exception as e:
+                    st.error(f"Error generating response: {str(e)}")
+
+    # Sidebar for PDF Upload
+    with st.sidebar:
+        st.title("Upload PDF 📂")
+        st.write("*This is for demonstration purposes. Do not submit any proprietary documents.*")
+        pdf_docs = st.file_uploader("Upload your PDF Files", accept_multiple_files=True)
+
+        if st.button("Process"):
+            if not pdf_docs:
+                st.error("Upload a PDF to start!")
+                return
+
+            with st.spinner("Processing, Chunking, and Caching..."):
+                raw_text = extract_pdf_text(pdf_docs)
+                text_chunks = split_text_into_chunks(raw_text)
+                create_and_save_vector_store(text_chunks)
+                st.success("Processing Done ✅")
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,9 @@
+streamlit
+langchain
+langchain-text-splitters
+langchain-community
+langchain-huggingface
+faiss-cpu
+PyPDF2
+sentence-transformers
+google-generativeai

src/config.py

@@ -0,0 +1,5 @@
+import os
+import google.generativeai as genai
+
+# Load API Key for Gemini
+genai.configure(api_key=os.getenv("GEMINI_API_KEY"))

src/pdf_processing.py

@@ -0,0 +1,20 @@
+from PyPDF2 import PdfReader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+def extract_pdf_text(pdf_docs):
+    """
+    Extracts text from a list of uploaded PDF files.
+    """
+    text = ""
+    for pdf in pdf_docs:
+        pdf_reader = PdfReader(pdf)
+        for page in pdf_reader.pages:
+            text += page.extract_text()
+    return text
+
+def split_text_into_chunks(text, chunk_size=10000, chunk_overlap=500):
+    """
+    Splits extracted text into smaller chunks for better processing.
+    """
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=chunk_size, chunk_overlap=chunk_overlap)
+    return text_splitter.split_text(text)

src/prompt_template.py

@@ -0,0 +1,19 @@
+from langchain.prompts import PromptTemplate
+
+def create_prompt_template():
+    """
+    Creates a structured prompt for querying the Gemini model.
+    """
+    prompt_template = """
+    Answer the question as detailed as possible from the provided context. 
+    If the answer contains structured data like tables or lists, respond in the same format. 
+    If the answer is not in the provided context, say, "The answer is not available in the context."
+
+    Context:
+    {context}
+
+    Question:
+    {question}
+    """
+
+    return PromptTemplate(template=prompt_template, input_variables=['context', 'question'])

src/query_handler.py

@@ -0,0 +1,23 @@
+import google.generativeai as genai
+from src.vector_store import load_vector_store
+from src.prompt_template import create_prompt_template
+
+def handle_user_query(user_question, index_name="faiss_index"):
+    """
+    Searches for relevant text in the vector store and generates a response using Gemini.
+    """
+    vector_store = load_vector_store(index_name)
+    docs = vector_store.similarity_search(user_question)
+
+    # Combine relevant document contents
+    context = "\n\n".join([doc.page_content for doc in docs])
+
+    # Format the prompt
+    prompt = create_prompt_template()
+    formatted_prompt = prompt.format(context=context, question=user_question)
+
+    # Generate response using Gemini AI
+    model = genai.GenerativeModel("gemini-1.5-flash")
+    response = model.generate_content(formatted_prompt)
+
+    return response.text if response.text else "No response generated."

src/vector_store.py

@@ -0,0 +1,18 @@
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+
+# Load Hugging Face embeddings model
+embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+
+def create_and_save_vector_store(text_chunks, index_name="faiss_index"):
+    """
+    Creates a FAISS vector store and saves it locally.
+    """
+    vector_store = FAISS.from_texts(text_chunks, embedding=embeddings)
+    vector_store.save_local(index_name)
+
+def load_vector_store(index_name="faiss_index"):
+    """
+    Loads the FAISS vector store.
+    """
+    return FAISS.load_local(index_name, embeddings, allow_dangerous_deserialization=True)