Shiao Wang

Results 7 comments of Shiao Wang

Have you solved this problem?I meet the same question now!

> > yes I'm only training one class "dog" of the lasot dataset, so where can I change the nu > > > See yaml file you used under experiment...

> See yaml file you used under experiment folder ![image](https://user-images.githubusercontent.com/37405764/201305134-d5e26cbd-d115-491e-87d0-a4b5e42e9971.png) 我在训练的时候一直卡在这个位置,也不报错,请问是什么原因啊? ![image](https://user-images.githubusercontent.com/109134157/201513252-e2374007-8a81-483a-8d1f-9df221dea696.png)

你好,能说一下是怎么改的吗?

您好! 这里的data['valid']可以不用管, return的data是采样的一对训练样本.

![Image](https://github.com/user-attachments/assets/283ca81c-fae7-44d5-820a-4b6b8c95eacf) 您好,感谢关注!您是用这一套参数跑的吗?

“max_num_voxels由16000改为40000”是转voxel数据需要的,我们这里用的是event frame。 > > ![Image](https://github.com/user-attachments/assets/283ca81c-fae7-44d5-820a-4b6b8c95eacf) 您好,感谢关注!您是用这一套参数跑的吗? > > 是的,除了batch_size我看你们论文里写的是32,就改成了32