Update app.py

app.py

@@ -95,14 +95,14 @@ model = AutoModelForCausalLM.from_pretrained(
     "m-a-p/YuE-s1-7B-anneal-en-cot",
     torch_dtype=torch.float16,
     attn_implementation="flash_attention_2",  # To enable flashattn, you have to install flash-attn
-)
-model.to(device)
+).to(device)
 model.eval()
 
 # Apply dynamic quantization
 model = torch.quantization.quantize_dynamic(
     model, {torch.nn.Linear}, dtype=torch.qint8
-)
+).to(device)
+model.eval()
 
 basic_model_config = './xcodec_mini_infer/final_ckpt/config.yaml'
 resume_path = './xcodec_mini_infer/final_ckpt/ckpt_00360000.pth'