Update app.py

app.py

@@ -1,5 +1,4 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
 from transformers import AutoTokenizer, AutoModelForCausalLM
 import torch
 import os
@@ -7,78 +6,41 @@ import os
 # Replace 'your_huggingface_token' with your actual Hugging Face access token
 access_token = os.getenv('token')
 
-# Initialize the tokenizer and model with the Hugging Face access token
+# Load the tokenizer and model
 tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b-it", use_auth_token=access_token)
 model = AutoModelForCausalLM.from_pretrained(
     "google/gemma-2b-it",
     torch_dtype=torch.bfloat16,
-    use_auth_token=access_token
+    device_map="auto",
+    use_auth_token=access_token# Automatically map to GPU if available
 )
-model.eval()  # Set the model to evaluation mode
 
-# Initialize the inference client (if needed for other API-based tasks)
-client = InferenceClient(provider="together", token=access_token)
-
-def conversation_predict(input_text):
-    """Generate a response for single-turn input using the model."""
-    # Tokenize the input text
-    input_ids = tokenizer(input_text, return_tensors="pt").input_ids
-
-    # Generate a response with the model
-    outputs = model.generate(input_ids, max_new_tokens=2048)
-
-    # Decode and return the generated response
-    return tokenizer.decode(outputs[0], skip_special_tokens=True)
-
-def respond(
-    message: str,
-    history: list[tuple[str, str]],
-    system_message: str,
-    max_tokens: int,
-    temperature: float,
-    top_p: float,
-):
-    """Generate a response for a multi-turn chat conversation."""
-    # Prepare the messages in the correct format for the API
-    messages = [{"role": "system", "content": system_message}]
-
-    for user_input, assistant_reply in history:
-        if user_input:
-            messages.append({"role": "user", "content": user_input})
-        if assistant_reply:
-            messages.append({"role": "assistant", "content": assistant_reply})
-
-    messages.append({"role": "user", "content": message})
-
-    # Get the complete response at once (no streaming)
-    response = client.chat_completion(
-        model="google/gemma-2b-it",
-        messages=messages,
-        max_tokens=max_tokens,
-        stream=False,
+# Define generation function
+def generate_text(prompt, max_tokens=200, temperature=0.7, top_p=0.9):
+    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+    outputs = model.generate(
+        **inputs,
+        max_new_tokens=max_tokens,
+        do_sample=True,
         temperature=temperature,
         top_p=top_p,
+        pad_token_id=tokenizer.eos_token_id
     )
-    
-    # Extract and return the full response
-    return response["choices"][0]["message"]["content"]
+    return tokenizer.decode(outputs[0], skip_special_tokens=True)
 
-# Create a Gradio ChatInterface demo
-demo = gr.ChatInterface(
-    fn=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)",
-        ),
+# Define Gradio interface
+interface = gr.Interface(
+    fn=generate_text,
+    inputs=[
+        gr.Textbox(label="Prompt", placeholder="Enter your prompt here...", lines=2),
+        gr.Slider(50, 512, value=200, label="Max Tokens"),
+        gr.Slider(0.1, 1.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(0.1, 1.0, value=0.9, step=0.05, label="Top-p (nucleus sampling)")
     ],
+    outputs=gr.Textbox(label="Generated Text"),
+    title="Gemma-2B Text Generator",
+    description="Enter a prompt and let Google's Gemma-2B-IT model generate a response."
 )
 
-if __name__ == "__main__":
-    demo.launch(share=True)
+# Launch the app
+interface.launch()