Long

Results 30 issues of Long

![image](https://user-images.githubusercontent.com/73996375/135072694-0c3214df-3eea-4791-a71a-c97215611fd3.png) Hello, I have a problem. What is the purpose of the identity transformation in the paper?

![image](https://user-images.githubusercontent.com/73996375/162610393-6f8de8e3-ad4b-42a0-b0d8-d8c590d29eaf.png) I want to use deep_supervision, but I find a error, should I how to deal with this error? What the function of the deep_supervision? Thank you!

### Bug description ![image](https://user-images.githubusercontent.com/73996375/178204623-e411b303-a57f-46df-893f-c00544f09055.png) I have got a error, we failed to use these method to obtain a results. I have got 4D matrix, but it needs to be 2D....

wontfix
module: methods
awaiting response

我设置了多个优化器去优化不同的任务,awl的学习率和优化器应该怎样选择呢?谢谢博主

![image](https://user-images.githubusercontent.com/73996375/122763859-66d27b00-d2d1-11eb-9b85-3ae4d708485e.png) 红框中的loss_1和2是取什么值呢?

怎么单独去做测试一些照片?

batch_size可以设置成其他值嘛?比如4,8,16等,为什么要是1呢

我的loss值有1-2万,但是最后的结果却是还行的,博主您最后的loss大概是多少呢?

博主您好,我想问一下伪标签的策略是怎么实现的?我看您提供的代码中并没有具体的实施

您好,您在划分影像和标签的时候,产生了有7个类别(0-6,Vaihingen数据集)。这意味着训练的时候应该按照7个类别来训练吗?但看您的论文上只是展示的图是前六个类别,但精度评价却是只提到了前五个类别,测试文件也是按照6个类别来进行的。我有个疑惑,这到底应该是按照几个类别来进行训练和预测呢?