leef

> [2018/12/13 09:22:03.066] INFO (15628) config : Load config from "config.json" > [2018/12/13 09:22:03.125] WARNING (15628) datasets.common : no image is valid > [2018/12/13 09:22:03.126] INFO (15628) **main** : Compute...

> > > [2018/12/13 09:22:03.066] INFO (15628) config : Load config from "config.json" > > > [2018/12/13 09:22:03.125] WARNING (15628) datasets.common : no image is valid > > > [2018/12/13...

>  > > @leaf918 > > In addition, mode=seg_tri training model, my loss curve is almost the same as you. Do you have some new opinions on this now?...

could you paste some result image here , @billzw

> @billzw hi~does the last black picture is your test output? yes.

I break it down. modify like this,works well. public abstract class BaseActivity extends AppCompatActivity

the script to handle annotation csv has ONLY the ability to annote one rectangle.

have you solved this issue? @arpu @JimmyChung208

Is there going to support the detection-like models in torchvision? There is some tests for torchvision's classification and segmentation.

> Finally, I succeeded. > step1: switch the two lines 97,98: > _loss_c = loss_c.view(num, -1) loss_c[pos] = 0 # filter out pos boxes for now_ > step2: change the...