Update app.py

app.py

@@ -1,28 +1,35 @@
 import gradio as gr
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
-import bitsandbytes as bnb
+from huggingface_hub import login, HfApi
 
-# Load the model and tokenizer with 8-bit quantization
-tokenizer = AutoTokenizer.from_pretrained("salmapm/llama2_salma")
-model = AutoModelForCausalLM.from_pretrained(
-    "salmapm/llama2_salma",
-    load_in_8bit=True,  # Enable 8-bit quantization
-    device_map='auto'  # Automatically maps model to available devices
-)
+def load_model(token):
+    # Log in with the user's token
+    login(token=token)
+
+    # Load the model and tokenizer
+    tokenizer = AutoTokenizer.from_pretrained("salmapm/llama2_salma")
+    model = AutoModelForCausalLM.from_pretrained(
+        "salmapm/llama2_salma",
+        load_in_8bit=True,  # Enable 8-bit quantization
+        device_map='auto'  # Automatically maps model to available devices
+    )
+
+    # Ensure the model is on the correct device (GPU if available)
+    device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+    model.to(device)
+
+    return model, tokenizer, device
+
+def respond(message, history, system_message, max_tokens, temperature, top_p, token):
+    if not token:
+        return "Please provide a Hugging Face token."
 
-# Ensure the model is on the correct device (GPU if available)
-device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-model.to(device)
+    try:
+        model, tokenizer, device = load_model(token)
+    except Exception as e:
+        return f"An error occurred: {e}"
 
-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:
@@ -33,7 +40,6 @@ def respond(
 
     messages.append({"role": "user", "content": message})
 
-    # Format the prompt for the model
     prompt = f"{system_message}\n" + "\n".join(
         [f"{msg['role']}: {msg['content']}" for msg in messages]
     )
@@ -48,22 +54,21 @@ def respond(
     )
 
     response = tokenizer.decode(outputs[0], skip_special_tokens=True)
-    yield response
+    return response
 
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
+# Create the Gradio interface
+demo = gr.Interface(
+    fn=respond,
+    inputs=[
+        gr.Textbox(label="Message"),
+        gr.Textbox(label="History (format: (user_message, assistant_response))", lines=2),
         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)",
-        ),
+        gr.Slider(minimum=0.1, maximum=1.0, value=0.95, step=0.05, label="Top-p (nucleus sampling)"),
+        gr.Textbox(label="Hugging Face Token", type="password")  # Token input field
     ],
+    outputs="text",
 )
 
 if __name__ == "__main__":