Cm xdg

hello @kikirizki ，how do you tansform the UA-DETRAC dataset to the format in JDE？I use the code in FairMOT(https://github.com/ifzhang/FairMOT/blob/master/src/gen_labels_15.py) and generate the label txts. The IDs in each txt file...

Thank you,I have addressed the proplem, but I meet a new error  I can't find any solution to this proplem,and I can't go next.

Thanks a lot! I can run it. I add the YOLOv3 dll to track a vedio, and I find a small bug in your code:  you didn't check the...

https://github.com/PaddlePaddle/PaddleOCR/blob/release/2.6/deploy/paddle2onnx/readme.md

onnxruntime和paddle inference的结果，耗时连小数点后面几位都一模一样吗，之前的版本这个文档，onnxruntime比paddleinference快很多  

Hello, have you solved this problem??

@LielinJiang 是的，直接安装的paddleGAN会出现我的问题，按您说的方法，安装develop版本就可以了，谢谢

谢谢，我把dec网络输入拆成6个了，现在可以正常导出和推理了

请问具体是怎么绕过呢？我在c++推理也同样出现了这个问题，不清楚修改哪里，是要改cu文件源码吗

@littletomatodonkey 你好，我这边测了3张图，后续耗时都是这样，而且cpu上c++版本推理比python慢特别多，可不可以以帮忙测一下，是只有我这边会这样吗？这种耗时正常还是不正常的，谢谢  附上另外2张图：