Update app.py

app.py

@@ -40,12 +40,14 @@ def init_model():
     model = LLaMa(args)
     torch.set_default_tensor_type(torch.FloatTensor)
     model = load_model(model, args.load_model_path)
-    print('load model done.')
     model.eval()
 
+    for name, parameter in model.named_parameters():
+        print(name)
+        print(parameter)
+    print(torch.cuda.max_memory_allocated() / 1024 ** 3)
     device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
     model.to(device)
-    print('to cuda.')
     lm_generation = LmGeneration(model, args.tokenizer)