persuelx
persuelx
@chen-311 the code have some problems,the results is not correct hog_list.append(hog(grey, mark, parameters)) the grey is Always zero
@nightfuryyy @kunyao2015 your problem has solved?I had the same problem when I trained my custom data。
But I have solved the bug,hhhh the accuracy is so high?
I can't remember some bugs, such as Tuple cannot be compared with int in >or
@Pay20Y Your answer is totally out of line with the requirements.You should see what people ask.
@PowerDi 请问后来你在Agx上这个int8的问题解决了吗？
@PowerDi 好的，谢谢，那用onnx 端对端推理的话，可以解决上述那个问题了么？
@khrlimam When I input two pictures, I can't get the result in your web app all the same.
@LCH1238 @missu123 你好请问这个问题解决了嘛？我是按照博主的环境配置的，只有torch，cuda版本有点区别。还有在执行dockerfile的时候出现了 => ERROR [internal] load metadata for nvcr.io/nvidia/tensorrt:22.07-py3 请问这个错误是什么原因呢
> You could refer to the [code](https://github.com/LCH1238/BEVDet/blob/export/tools/export/export_datayaml.py) to generate data_infos for nusence-mini dataset. Before this, you need to prepare a pkl file similar to "bevdetv2-nuscenes_infos_val.pkl". 请问你生成data_info的时候使用的nuscenes mini还是全部的数据集，你的配置文件中的图像文件在mini中不存在？ 谢谢了