Li

> Hi, for online recognition, we need to use a uni-directional TSM, but we have not included the code. But it would be easy to modify by yourself (just change...

> 于老师,您好　 　　我在使用项目中提供的训练脚本得到的模型性能比您开源的模型性能要差1~1.5个点.请问开源模型是否在训练时进行了别的参数调优,还是仅仅是因为数据shuffle不同导致的? 此外,还想请教一下,如果我们不需要输出人脸关键点预测,那么在训练时删除人脸关键点,对于人脸检测性能会有帮助吗? 您好，请问您使用什么样的配置参数呢？ “yufacedet.taml” 里的原参数吗？ 训练1000个epochs？

hello, I'd like to compare with this paper,but i find a question.The batch size in your paper is 256, but the configs file is 128, so which is right? I'm...

hi, I'd like to use the best results, it can be used for testing,but I want to use it for pretraining, i can't load it.Can you give me some suggests?...

This is the error if i reload the chekpoint to pretrain. Traceback (most recent call last): File "D:/Myselfs/Codes/Pycharm/CC/train.py", line 129, in optimizer.load_state_dict(checkpoint['optimizer']) KeyError: 'optimizer'

thank you, it worked

I set the queue size and the dataset size to be consistent , but I have got a worse result.

@hejing-maker 您好，请问数据标注是怎样标注呢？您qq拒绝添加~

> Hi, > > I had a problem after training the first epoch. It looks like the label of cv data are all zero. Is there anything wrong when labeling...

> Hey, > > > I had a problem after training the first epoch. It looks like the label of cv data are all zero. Is there anything wrong when...