Hao Zhang

> @Github3G ,您好，请问由HRNet生成的pred.mat（7247,16,2）经过[(https://github.com/anibali/eval-mpii-pose](url)转换后，生成的.mat文件就可以上传到官网了么？期待您的回复， 您好，我也遇到了同样的问题，请问你解决了么

> @leoxiaobin > Hi, I made a PR for yolov3-HRnet, however something wired. I use two ways. > > > ONE, I get dt_boxes from yolo then `python tools/test.py TEST.USE_GT_BBOX...

> Hello.Thank you for sharing your code. I have test the model file traind on MPII train dataset the repo provide on MPII valid dataset. But I got different results...

> > Hello.Thank you for sharing your code. I have test the model file traind on MPII train dataset the repo provide on MPII valid dataset. But I got different...

> I finally get 90.3. I change the learning strategy. The total epoch is 140, and I change the learning rate at 90 and 120 epoch. How many GPUs and...

Have you solved the problem? I'm in the same situation. On the test set, validation set, and training set, the results are low and close to yours.