Update app.py

app.py

@@ -72,7 +72,7 @@ device = "cuda:0"
 # Load model and tokenizer outside the generation function (load once)
 print("Loading model...")
 model = AutoModelForCausalLM.from_pretrained(
-    "m-a-p/YuE-s1-7B-anneal-en-cot",  # "m-a-p/YuE-s1-7B-anneal-en-icl",
+    "m-a-p/YuE-s1-7B-anneal-en-icl", # "m-a-p/YuE-s1-7B-anneal-en-cot", 
     torch_dtype=torch.float16,
     attn_implementation="flash_attention_2",
 ).to(device)
@@ -220,7 +220,7 @@ def generate_music(
                     logits_processor=LogitsProcessorList([BlockTokenRangeProcessor(0, 32002), BlockTokenRangeProcessor(32016, 32016)]),
                     guidance_scale=guidance_scale,
                     use_cache=True,
-                    num_beams=1
+                    num_beams=3
                 )
                 if output_seq[0][-1].item() != mmtokenizer.eoa:
                     tensor_eoa = torch.as_tensor([[mmtokenizer.eoa]]).to(model.device)