Xudong Wang

I could have succeeded by fixing the input, but failed with dynamic input

I didn't use torch.where and nonzero in my code, but I still had the same error!! Why??

I run into the same problem,

> 抱歉，但是当我进行安装时，我无法成功安装mamba-ssm，我遇到了以下问题：我尝试了pip install和conda install，并且还尝试了pip install。从这个存储库中，所有这些都无法工作。 This is because the cuda version is too low and requires cuda 11.6 and above.

In some cases, the classification of LB1+2, LB3, LB4, LB5 and LB6, LB7+8, LB9, LB10 is wrong and should be in the upper lobe of the class to the lower...

> I am using windows, where the local variables are used second, and the npz file in stage 01 is converted by its own method. In the second step (airway/image_processing/remove_all_0_layers.py),...

> 可以先用这个脚本测一下导出的结果 > > https://github.com/lyuwenyu/RT-DETR/blob/main/rtdetrv2_pytorch/references/deploy/rtdetrv2_tensorrt.py 有测试，无论是官方提供的权重文件还是我自己的权重文件，都出向没有“create_execution_context”这个模块，未定义

> trt版本是多少 > > box不对看下是不是数据类型的问题 trt版本是8.6.1； 但是标签和得分都是对的