Initial commit: PDF RAG chatbot with LangChain and Groq

.gitignore

@@ -0,0 +1,51 @@
+# Python
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+
+# Virtual environments
+venv/
+env/
+ENV/
+.venv
+
+# IDEs
+.vscode/
+.idea/
+*.swp
+*.swo
+*~
+
+# Gradio
+gradio_queue.db
+flagged/
+
+# Vector stores
+faiss_index/
+chroma_db/
+
+# Temporary files
+*.tmp
+*.log
+.DS_Store
+
+# Environment variables
+.env
+.env.local

README.md

@@ -0,0 +1,98 @@
+# 📄 PDF RAG Chatbot
+
+A Retrieval Augmented Generation (RAG) chatbot that allows you to upload PDF documents and have conversations based on their content using AI.
+
+## Features
+
+- 📤 **PDF Upload**: Upload any PDF document
+- 🤖 **AI-Powered Chat**: Ask questions about your PDF content
+- 🔍 **Semantic Search**: Uses vector embeddings to find relevant information
+- 💬 **Conversation Memory**: Maintains context throughout the conversation
+- 🚀 **Fast Processing**: Powered by Groq's LLM API
+
+## Tech Stack
+
+- **UI**: Gradio
+- **LLM**: Groq (Llama 3.3 70B)
+- **Framework**: LangChain
+- **Embeddings**: HuggingFace (all-MiniLM-L6-v2)
+- **Vector Store**: FAISS
+- **PDF Processing**: PyPDF
+
+## Installation
+
+1. Clone the repository:
+```bash
+git clone <your-repo-url>
+cd pdf-rag-chatbot
+```
+
+2. Install dependencies:
+```bash
+pip install -r requirements.txt
+```
+
+3. Set up your Groq API key:
+```bash
+export GROQ_API_KEY="your-api-key-here"
+```
+
+Get your free API key from [Groq Console](https://console.groq.com)
+
+## Usage
+
+1. Run the application:
+```bash
+python app.py
+```
+
+2. Open your browser to the displayed URL (usually `http://localhost:7860`)
+
+3. Upload a PDF file
+
+4. Wait for processing to complete
+
+5. Start asking questions about the PDF content!
+
+## How It Works
+
+1. **PDF Processing**: The uploaded PDF is split into smaller chunks
+2. **Embedding**: Each chunk is converted into a vector embedding
+3. **Vector Storage**: Embeddings are stored in FAISS for fast retrieval
+4. **Query**: When you ask a question, the system finds the most relevant chunks
+5. **Response**: The LLM generates an answer based on the retrieved context
+
+## Environment Variables
+
+- `GROQ_API_KEY`: Your Groq API key (required)
+
+## Deployment
+
+### Hugging Face Spaces
+
+1. Create a new Space on Hugging Face
+2. Upload all files
+3. Add `GROQ_API_KEY` to Space secrets
+4. Your app will be live!
+
+### Local Development
+
+```bash
+python app.py
+```
+
+## Example Questions
+
+After uploading a PDF, you can ask questions like:
+- "What is the main topic of this document?"
+- "Summarize the key points"
+- "What does the document say about [specific topic]?"
+- "Can you explain [concept] from the document?"
+
+## License
+
+MIT License
+
+## Contributing
+
+Contributions are welcome! Please feel free to submit a Pull Request.

app.py

@@ -0,0 +1,219 @@
+import os
+import gradio as gr
+from langchain_groq import ChatGroq
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import FAISS
+from langchain_community.embeddings import HuggingFaceEmbeddings
+from langchain.chains import ConversationalRetrievalChain
+from langchain.memory import ConversationBufferMemory
+from langchain_community.document_loaders import PyPDFLoader
+import tempfile
+import shutil
+
+MODEL_NAME = "llama-3.3-70b-versatile"
+DEFAULT_API_KEY = os.getenv("GROQ_API_KEY", "")
+
+# Global variables
+vectorstore = None
+conversation_chain = None
+chat_history = []
+
+def process_pdf(pdf_file, api_key):
+    """Process uploaded PDF and create vector store"""
+    global vectorstore, conversation_chain, chat_history
+    
+    if not api_key:
+        return "Please provide a Groq API key first.", None
+    
+    if pdf_file is None:
+        return "Please upload a PDF file.", None
+    
+    try:
+        # Save uploaded file temporarily
+        temp_dir = tempfile.mkdtemp()
+        temp_pdf_path = os.path.join(temp_dir, "uploaded.pdf")
+        shutil.copy(pdf_file.name, temp_pdf_path)
+        
+        # Load PDF
+        loader = PyPDFLoader(temp_pdf_path)
+        documents = loader.load()
+        
+        # Split documents into chunks
+        text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=1000,
+            chunk_overlap=200,
+            length_function=len
+        )
+        chunks = text_splitter.split_documents(documents)
+        
+        # Create embeddings and vector store
+        embeddings = HuggingFaceEmbeddings(
+            model_name="sentence-transformers/all-MiniLM-L6-v2"
+        )
+        vectorstore = FAISS.from_documents(chunks, embeddings)
+        
+        # Initialize LLM
+        llm = ChatGroq(
+            groq_api_key=api_key,
+            model_name=MODEL_NAME,
+            temperature=0.7,
+            max_tokens=1024
+        )
+        
+        # Create conversation chain
+        memory = ConversationBufferMemory(
+            memory_key="chat_history",
+            return_messages=True,
+            output_key="answer"
+        )
+        
+        conversation_chain = ConversationalRetrievalChain.from_llm(
+            llm=llm,
+            retriever=vectorstore.as_retriever(search_kwargs={"k": 3}),
+            memory=memory,
+            return_source_documents=True
+        )
+        
+        # Reset chat history
+        chat_history = []
+        
+        # Cleanup
+        shutil.rmtree(temp_dir)
+        
+        return f"✅ PDF processed successfully! Found {len(chunks)} text chunks. You can now ask questions about the document.", []
+        
+    except Exception as e:
+        return f"Error processing PDF: {str(e)}", None
+
+def chat_with_pdf(message, chat_history_ui, api_key):
+    """Handle chat interactions with the PDF content"""
+    global conversation_chain, chat_history
+    
+    if not message.strip():
+        return chat_history_ui, ""
+    
+    if conversation_chain is None:
+        chat_history_ui.append({
+            "role": "user",
+            "content": message
+        })
+        chat_history_ui.append({
+            "role": "assistant",
+            "content": "Please upload a PDF file first before asking questions."
+        })
+        return chat_history_ui, ""
+    
+    try:
+        # Add user message
+        chat_history_ui.append({
+            "role": "user",
+            "content": message
+        })
+        
+        # Get response from RAG chain
+        response = conversation_chain({"question": message})
+        answer = response["answer"]
+        
+        # Add assistant response
+        chat_history_ui.append({
+            "role": "assistant",
+            "content": answer
+        })
+        
+        return chat_history_ui, ""
+        
+    except Exception as e:
+        chat_history_ui.append({
+            "role": "assistant",
+            "content": f"Error: {str(e)}"
+        })
+        return chat_history_ui, ""
+
+def reset_chat():
+    """Reset the conversation"""
+    global conversation_chain, vectorstore, chat_history
+    conversation_chain = None
+    vectorstore = None
+    chat_history = []
+    return [], "Ready to upload a new PDF."
+
+# Build Gradio Interface
+with gr.Blocks(title="PDF RAG Chatbot") as demo:
+    gr.Markdown("# 📄 PDF RAG Chatbot")
+    gr.Markdown("Upload a PDF and chat with its content using AI")
+    gr.Markdown(f"**Model:** `{MODEL_NAME}`")
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            if not DEFAULT_API_KEY:
+                api_key_input = gr.Textbox(
+                    label="Groq API Key",
+                    placeholder="Enter your Groq API key here...",
+                    type="password"
+                )
+            else:
+                api_key_input = gr.Textbox(
+                    type="password",
+                    value=DEFAULT_API_KEY,
+                    visible=False
+                )
+            
+            pdf_upload = gr.File(
+                label="Upload PDF",
+                file_types=[".pdf"],
+                type="filepath"
+            )
+            
+            process_btn = gr.Button("Process PDF", variant="primary")
+            status_text = gr.Textbox(
+                label="Status",
+                value="Upload a PDF to get started.",
+                interactive=False
+            )
+            
+            clear_btn = gr.Button("Reset Chat", variant="stop")
+        
+        with gr.Column(scale=2):
+            chatbot = gr.Chatbot(height=500)
+            
+            with gr.Row():
+                msg = gr.Textbox(
+                    label="Message",
+                    placeholder="Ask a question about the PDF...",
+                    scale=4
+                )
+                submit_btn = gr.Button("Send", scale=1)
+    
+    if not DEFAULT_API_KEY:
+        gr.Markdown("### Instructions:")
+        gr.Markdown("1. Get a free API key from [Groq Console](https://console.groq.com)")
+        gr.Markdown("2. Enter your API key above")
+        gr.Markdown("3. Upload a PDF file")
+        gr.Markdown("4. Ask questions about the content!")
+    
+    # Event handlers
+    process_btn.click(
+        process_pdf,
+        inputs=[pdf_upload, api_key_input],
+        outputs=[status_text, chatbot]
+    )
+    
+    msg.submit(
+        chat_with_pdf,
+        inputs=[msg, chatbot, api_key_input],
+        outputs=[chatbot, msg]
+    )
+    
+    submit_btn.click(
+        chat_with_pdf,
+        inputs=[msg, chatbot, api_key_input],
+        outputs=[chatbot, msg]
+    )
+    
+    clear_btn.click(
+        reset_chat,
+        outputs=[chatbot, status_text]
+    )
+
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -0,0 +1,9 @@
+gradio==4.44.0
+langchain==0.3.7
+langchain-groq==0.2.1
+langchain-community==0.3.5
+pypdf==5.1.0
+sentence-transformers==3.3.1
+faiss-cpu==1.9.0
+transformers==4.46.3
+torch==2.5.1