smallsun1122

同学，你好，时间太久了，我有点忘记了，我在网上找了一些解决办法，看能不能帮到你，这些解决办法我看着有点熟悉 ，可能是之前用过的，希望可以帮到你。 ------------------ 原始邮件 ------------------ 发件人: "Tencent/FaceDetection-DSFD" ***@***.***>; 发送时间: 2021年4月19日(星期一) 晚上8:27 ***@***.***>; 抄送: "small ***@***.******@***.***>; 主题: Re: [Tencent/FaceDetection-DSFD] The following error was reported during my running demo.py: RuntimeError: Legacy autograd function with non-static forward method...

@lijiannuist how to train the mode ?

When I run demo.py, I need to specify eval_tools. Is this eval_tools downloaded from the Internet? Or is it an empty folder I created?

Excuse me, this program does not seem to specify the gpu to run, is it run with cpu?

Excuse me, have you solved it? I have the same problem

Excuse me, have you solved it? I have the same problem

Excuse me, have you solved it? I have the same problem

> 您好，我将您的开源代码clone下来以后，试图运行train.py进行尝试，但是似乎在dataiter.py的DsfdDataIter类下出现了些问题，首先是未定义该类的__len__方法，我试着添加了该方法后，在DataIter类下build_iter方法下的ds = MultiProcessPrefetchData(ds, self.prefetch_size, self.process_num)语句又发生了进程间冲突的报错，因此想请教是代码本身存在问题还是我个人操作出现了错误，谢谢~ 请问你解决了吗？我也出现了问题，但问题是DsfdDataIter have no len

您好，请问一下，您这边在训练的过程中利用visdom显示的loss可以正常显示吗？我这边利用visdom显示loss的时候刚刚开始是有数据的，但是过一会数据就不会变化，然后刷新一下浏览器，图就没有了。您遇到过这种情况吗？

I now have similar doubts, the training process seems to use the 2007VOC and 2012VOC, why the eval.py used in the 2007VOC?