Create app.py

app.py

@@ -0,0 +1,130 @@
+import os
+import gradio as gr
+import numpy as np
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.document_loaders import DirectoryLoader, TextLoader
+from langchain.embeddings import HuggingFaceEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.chains import ConversationalRetrievalChain
+from langchain.memory import ConversationBufferMemory
+from langchain.llms import HuggingFaceHub
+
+# Set up environment variables for HuggingFace
+os.environ["HUGGINGFACEHUB_API_TOKEN"] = os.getenv("HUGGINGFACE_API_TOKEN")  # Replace with your token
+
+# Create a directory for document storage if it doesn't exist
+os.makedirs("documents", exist_ok=True)
+
+# Function to load documents
+def load_documents(directory="documents"):
+    loader = DirectoryLoader(directory, glob="**/*.txt", loader_cls=TextLoader)
+    documents = loader.load()
+    return documents
+
+# Function to process documents and create vector store
+def process_documents():
+    documents = load_documents()
+    
+    # Split documents into chunks
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=1000,
+        chunk_overlap=200
+    )
+    chunks = text_splitter.split_documents(documents)
+    
+    # Create embeddings
+    embeddings = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+    
+    # Create vector store
+    vector_store = FAISS.from_documents(chunks, embeddings)
+    
+    return vector_store
+
+# Create RAG chain
+def create_chain(vector_store):
+    # Initialize the LLM
+    llm = HuggingFaceHub(
+        repo_id="google/flan-t5-large", 
+        model_kwargs={"temperature": 0.5, "max_length": 512}
+    )
+    
+    # Create memory for the conversation
+    memory = ConversationBufferMemory(
+        memory_key="chat_history",
+        return_messages=True
+    )
+    
+    # Create the conversational chain
+    chain = ConversationalRetrievalChain.from_llm(
+        llm=llm,
+        retriever=vector_store.as_retriever(search_kwargs={"k": 3}),
+        memory=memory
+    )
+    
+    return chain
+
+# Initialize variables for handling chat state
+vector_store = None
+chain = None
+chat_history = []
+
+# Function to handle file uploads
+def upload_file(files):
+    for file in files:
+        file_path = os.path.join("documents", os.path.basename(file.name))
+        with open(file_path, "wb") as f:
+            f.write(file.read())
+    
+    global vector_store, chain
+    vector_store = process_documents()
+    chain = create_chain(vector_store)
+    
+    return "Files uploaded and processed successfully!"
+
+# Function to handle user queries
+def chat(message, history):
+    global chain, chat_history, vector_store
+    
+    # Initialize vector store and chain if they don't exist
+    if vector_store is None:
+        if os.path.exists("documents") and any(os.path.isfile(os.path.join("documents", f)) for f in os.listdir("documents")):
+            vector_store = process_documents()
+            chain = create_chain(vector_store)
+        else:
+            return "Please upload documents first to initialize the chatbot."
+    
+    # Convert Gradio history to the format expected by the chain
+    if history:
+        chat_history = [(turn[0], turn[1]) for turn in history]
+    
+    # Get response from chain
+    response = chain({"question": message})
+    
+    return response['answer']
+
+# Create Gradio interface
+with gr.Blocks(title="RAG Chatbot") as demo:
+    gr.Markdown("# RAG-based Conversational Chatbot")
+    gr.Markdown("Upload text documents and chat with an AI that can answer questions based on their content.")
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            file_output = gr.Textbox(label="Upload Status")
+            file_input = gr.File(
+                file_count="multiple",
+                label="Upload Documents (.txt files)"
+            )
+            upload_button = gr.Button("Process Documents")
+            upload_button.click(upload_file, inputs=[file_input], outputs=[file_output])
+        
+        with gr.Column(scale=2):
+            chatbot = gr.Chatbot(height=400)
+            msg = gr.Textbox(label="Ask a question about your documents")
+            
+            msg.submit(chat, inputs=[msg, chatbot], outputs=[chatbot])
+            clear = gr.Button("Clear")
+            clear.click(lambda: [], outputs=[chatbot])
+
+# Launch the app
+if __name__ == "__main__":
+    demo.launch()