app.py

import gradio as gr
from huggingface_hub import InferenceClient
import chromadb
from chromadb.config import Settings

from chromadb import PersistentClient

# Initialize the inference client with model
inference_client = InferenceClient(model="unsloth/Llama-3.2-3B-Instruct") 

# path to the ChromaDB directory
client_db = PersistentClient(path="./chromadb_directory/chromadb_file")

# Load collection
collection = client_db.get_collection("my_collection")

# Function to retrieve documents from ChromaDB, ensuring results are strings
def retrieve_from_chromadb(query):
    results = collection.query(query_texts=query, n_results=5)  # Adjust n_results as needed
    # Ensure each document is a string
    documents = [str(doc) for doc in results['documents']]
    return documents

# Respond function for the chatbot
def respond(
    message,
    history: list[tuple[str, str]],
    system_message,
    max_tokens,
    temperature,
    top_p,
):
    # Prepare messages for the model
    messages = [{"role": "system", "content": system_message}]

    # Add conversation history
    for val in history:
        if val[0]:
            messages.append({"role": "user", "content": val[0]})
        if val[1]:
            messages.append({"role": "assistant", "content": val[1]})

    # Retrieve relevant documents from ChromaDB
    retrieved_docs = retrieve_from_chromadb(message)
    
    # Join the documents to create a context for the user query
    context = "\n".join(retrieved_docs) + "\nUser: " + message
    messages.append({"role": "user", "content": context})

    response = ""

    # Generate response using the Inference Client
    for message in inference_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

# Gradio Chat Interface
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)",
        ),
    ],
)

if __name__ == "__main__":
    demo.launch()