little-seasalt
little-seasalt
> > > 您好，请问本论文的精度怎么才能达到论文展示的呀，我UCF-QNRF共1500轮在1315轮达到最好，test出来MAE:83.4，MSE:146。与论文展示的77.3和131.5还有不少差距，最好模型大小154.19M，与您提供的218.36M大小也不相同，请问这是什么原因呀？JHU的精度也没有达到论文展示的。 > > > > > > 你好，可以问下你jhu的测试结果吗，我在jhu上训练时发现模型难以收敛，不知道是什么原因，训练日志如下  > >  我当时收敛很正常呀，JHU 在 480 就达到最好精度了 作者您好，我想请问一下，我在训练过程中发现我的验证轮次时长大约是训练轮次时长的3倍，请问这大概是因为什么呢？不知您能否帮忙解答一下？
> Here are a few things you may check before training: > > * **Data preprocessing:** > Please ensure that the images and corresponding ground-truth points are processed properly. For...
> Here are a few things you may check before training: > > * **Data preprocessing:** > Please ensure that the images and corresponding ground-truth points are processed properly. For...
> Sorry for the late reply. We notice that the performance on the UCF-QNRF dataset seems somewhat behind the reported results. We will try to figure out the reasons and...
> We found that abnormal loss values computed on dense images could affect the performance. A simple workaround is to eliminate such abnormal losses during training. What modifications should be...
> > > We found that abnormal loss values computed on dense images could affect the performance. A simple workaround is to eliminate such abnormal losses during training. > >...
> > > We found that abnormal loss values computed on dense images could affect the performance. A simple workaround is to eliminate such abnormal losses during training. > >...
> Did you modify `img_ds_idx = den_sort[len(den_sort)//2]` to `img_ds_idx = den_sort[1:len(den_sort)//2]` ? In addition, it is recommended to increase the number of training samples in dataloader by a proportion of...
> > > > > We found that abnormal loss values computed on dense images could affect the performance. A simple workaround is to eliminate such abnormal losses during training....
> @cxliu0 Thanks the author for providing precious information 👍 . However, it seems undeterministic to attain the same results reported in the paper. I got also 96.1 MAE on...