Update app.py

app.py

@@ -1,11 +1,43 @@
-from document_loader import load_html_documents
+import gradio as gr
+from langchain.chains import RetrievalQA
+from langchain.llms import DeepseekLLM  # Or your preferred LLM
+from langchain.vectorstores import Chroma
+from langchain.embeddings import HuggingFaceEmbeddings
+import os
+import preprocess  # Import the preprocess module
+import create_database  # Import the create_database module
 
-if __name__ == "__main__":
-    print("Starting Dubai Law Chatbot...")
 
-    print("Loading documents...")
-    documents = load_html_documents('./data')
-    print(f"Loaded {len(documents)} documents.")
+# --- Preprocessing and Database Creation ---
 
+# Preprocess data if not already done
+if not os.path.exists("db"):  # Check if database exists
+    preprocess.preprocess_and_save("./documents", "preprocessed_data.json")  # Update path
+    create_database.create_vector_database("preprocessed_data.json", "db")
 
+# --- RAG Pipeline ---
 
+# Load the vector database
+embedding_model = HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2")
+vector_db = Chroma(persist_directory="db", embedding_function=embedding_model)
+retriever = vector_db.as_retriever(search_kwargs={"k": 3})
+
+# Load your LLM
+llm = DeepseekLLM(model_name="deepseek-ai/deepseek-coder-7b-instruct")  # Update if needed
+
+# Create the RetrievalQA chain
+qa_chain = RetrievalQA(llm=llm, retriever=retriever)
+
+# --- Gradio Interface ---
+
+def chatbot_interface(question):
+    return qa_chain.run(question)
+
+iface = gr.Interface(
+    fn=chatbot_interface,
+    inputs="text",
+    outputs="text",
+    title="Dubai Legislation AI Chatbot"
+)
+
+iface.launch()