2.6版本使用jupyter notebook运行程序后，显存被占用，再次运行程序显示：

[xiaofeifei-1]

bug描述 Describe the Bug

2.6版本使用jupyter notebook运行程序后，显存被占用，再次运行程序显示：

其他补充信息 Additional Supplementary Information

---

MemoryError Traceback (most recent call last)
Cell In[3], line 131
127 train(args)
130 if name == 'main':
--> 131 launch()
132 pass

Cell In[3], line 127
124 self.lr = 1.5e-4
126 args = ARGS()
--> 127 train(args)

Cell In[3], line 89
87 t = diffusion.sample_timesteps(images[0].shape[0])
88 x_t, noise = diffusion.noise_images(images[0], t)
---> 89 predicted_noise = model(x_t, t)
90 loss = mse(noise, predicted_noise) # 损失函数
92 opt.clear_grad()

File ~/anaconda3/envs/paddle_2.6/lib/python3.8/site-packages/paddle/nn/layer/layers.py:1429, in Layer.call(self, *inputs, **kwargs)
1420 if (
1421 (not in_to_static_mode())
1422 and (not self._forward_pre_hooks)
(...)
...
Please check whether there is any other process using GPU 0.

1. If yes, please stop them, or start PaddlePaddle on another GPU.
2. If no, please decrease the batch size of your model.
   (at ../paddle/fluid/memory/allocation/cuda_allocator.cc:86)

[westfish]

restart一下kernel再跑跑看呢