Update app.py

app.py

@@ -1,15 +1,13 @@
-import gradio as gr
-from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
-from accelerate import init_empty_weights
 from transformers import AutoModelForCausalLM, AutoTokenizer
+import gradio as gr
+
+# Vérifiez si un GPU est disponible avec ZeroGPU
+device = "cuda" if torch.cuda.is_available() else "cpu"
 
 # Charger le modèle
 model_name = "bigcode/starcoder2-15b-instruct-v0.1"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
-
-# Initialisation conditionnelle
-device = "cuda" if torch.cuda.is_available() else "cpu"
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype=torch.float16 if device == "cuda" else torch.float32
@@ -21,7 +19,6 @@ def generate_text(prompt):
     outputs = model.generate(inputs["input_ids"], max_length=200)
     return tokenizer.decode(outputs[0], skip_special_tokens=True)
 
-
 # Interface utilisateur Gradio
 interface = gr.Interface(
     fn=generate_text,