tensorboy
tensorboy
> ``` > INFO:torch.distributed.elastic.multiprocessing:Setting worker0 reply file to: /tmp/torchelastic_zg87tjfx/none_hh2up_uh/attempt_0/0/error.json > INFO:torch.distributed.elastic.multiprocessing:Setting worker1 reply file to: /tmp/torchelastic_zg87tjfx/none_hh2up_uh/attempt_0/1/error.json > INFO:torch.distributed.elastic.multiprocessing:Setting worker2 reply file to: /tmp/torchelastic_zg87tjfx/none_hh2up_uh/attempt_0/2/error.json > INFO:torch.distributed.elastic.multiprocessing:Setting worker3 reply file to: /tmp/torchelastic_zg87tjfx/none_hh2up_uh/attempt_0/3/error.json...
I will for sure come back to let you know if I can solve it.
明天回答你可以吗，因为只有2693张里面有人。 On Sun, Mar 8, 2020 at 10:45 PM Dantju wrote: > @tensorboy > 请问为什么evalute中的载入的val数量只有2693，coco val中总共有5000才对 > > — > You are receiving this because you were mentioned. > Reply...
@Dantju 你好，我其实并没有太多的改进，就是调了调参数，DLA34对应的参数都在这里：https://github.com/tensorboy/centerpose/blob/master/experiments/dla_34_512x512.yaml
当然有了，只要不断调就会越来越高。 On Mon, Mar 9, 2020 at 10:42 PM Dantju wrote: > @tensorboy 调了参数之后AP有上升吗 > > — > You are receiving this because you were mentioned. > Reply to this...
如果都枚举出来的话，其实我也忘的差不多了。。。反正能看到不少提高的地方。。。
PE和PD是什么的缩写呢。@Dantju
Will let you know when I found the problem.
I didn't have time to tune it now, will let you know if I have the result.
Yes, the limb_width is 1 in default, I tried to set it to more than 1 (1.1, 1.2, 1.3), it shows a slight increase in the mAP.