Legal Query chatbot intial commit

app.py

@@ -1,63 +1,139 @@
+import os
+import openai
+from langchain_community.embeddings import OpenAIEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain_community.llms import OpenAI
+from langchain.chains import ConversationChain
+from langchain_community.document_loaders import PyPDFLoader
+from langchain.memory import ConversationBufferMemory
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_openai import ChatOpenAI
+from langchain.memory import ConversationSummaryMemory
+
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
-
-
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
+from PyPDF2 import PdfReader
+from langchain.agents import initialize_agent, Tool
+from langchain.schema import HumanMessage, AIMessage
+from langchain_core.exceptions import OutputParserException
+
+# Set up OpenAI API key
+#openai.api_key = os.getenv("OPENAI_API_KEY")
+
+#TODO: Need to remove
+apiKey = open("key.txt", "r").readline().strip('\n') 
+
+
+
+# Load PDF
+def read_pdf(file_paths):
+    combined_text = ""
+    for file_path in file_paths:
+        with open(file_path, "rb") as file:
+            reader = PdfReader(file)
+            text = ""
+            for page in reader.pages:
+                text += page.extract_text()
+            combined_text += text + "\n\n"  # Add a newline for separation between files
+    return combined_text
+
+# Load legal document (Constitution of India)
+pdf_file_path = ["property_law.pdf","ipc.pdf","constitution_of_india.pdf"]
+document_text = read_pdf(pdf_file_path)
+
+# Split the text into smaller chunks (e.g., 1,000 characters each)
+text_splitter = RecursiveCharacterTextSplitter(
+    chunk_size=1000,  # Adjust this size based on your needs
+    chunk_overlap=100  # To keep some overlap between chunks for better context
+)
+
+# Split the document text
+chunks = text_splitter.split_text(document_text)
+
+# Initialize embeddings and FAISS vector store
+embeddings = OpenAIEmbeddings(openai_api_key=apiKey)
+vector_db = FAISS.from_texts(chunks, embeddings)
+
+
+
+
+# Function to retrieve relevant content from vector DB
+def retrieve_from_db(query):
+    results = vector_db.similarity_search(query, k=1)
+    return results[0].page_content
+
+
+# Initialize OpenAI LLM
+llm = ChatOpenAI(openai_api_key=apiKey)
+
+
+# Define agent tools
+tools = [
+    Tool(
+        name="DocumentRetriever",
+        func=retrieve_from_db,
+        description="Retrieves relevant legal information from pre-loaded documents"
+    )
+]
+
+
+# Initialize memory and agent
+memory = ConversationSummaryMemory(llm=llm)
+agent = initialize_agent(
+    tools=tools,
+    agent_type="zero-shot-react-description",
+    llm=llm,
+    memory=memory,
+    handle_parsing_errors=True
 )
 
 
+# Function to interact with the agent and store conversation
+def chatbot(input_text, chat_history):
+    try:    
+        
+        # Run the agent with the input and memory history
+        response = agent.run(input_text)
+
+        # Check if the response is "N/A" and replace with a custom message
+        if response == "N/A":
+            response = "Sorry, I couldn't understand your question. Please ask a specific question regarding IPC, Transfer of Property and Constitution of India."
+        
+        # Store the assistant's response in memory
+        memory.save_context({"user": input_text}, {"assistant": response})
+
+        # Update chat history with the new response
+        chat_history.append([input_text, response])
+    
+        
+        return chat_history
+    
+    except OutputParserException as e:
+        # Handle the exception and notify the user
+        error_message = "Sorry, I couldn't understand your question. Please ask a specific question regarding IPC, Transfer of Property and Constitution of India."
+        # Append the error message to chat history
+        chat_history.append([error_message, input_text])
+        print("Error:", str(e))
+
+        return chat_history
+
+# Gradio UI
+def gradio_interface():
+    with gr.Blocks() as demo:
+        gr.Markdown("# Legal Query Chatbot")
+
+         # Create chat UI with custom class
+        with gr.Column():
+            chatbot_ui = gr.Chatbot()
+            user_input = gr.Textbox(show_label=True, placeholder="Enter your INDIAN PENAL CODE, TRANSFER OF PROPERTY, CONSTITUTION OF INDIA query here...")
+            submit_button = gr.Button("Submit")
+            submit_button.click(fn=chatbot, inputs=[user_input, chatbot_ui], outputs=chatbot_ui)
+            user_input.submit(fn=chatbot, inputs=[user_input, chatbot_ui], outputs=chatbot_ui)
+
+    return demo
+
+
+# Run the app
+app = gradio_interface()
+
 if __name__ == "__main__":
-    demo.launch()
+    app.launch()

requirements.txt

@@ -1 +1,5 @@
-huggingface_hub==0.22.2
+openai
+langchain
+langchain_community
+gradio
+PyPDF2