cyizhuo

I think 'reset_parameters()' can solve nan problem

Don't know if it helps or not, but I created conda env with: conda install scipy==1.0.0 (must be installed first) conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch and Net-Dissection-Lite worked...

Hello friend, did you sloved this problem? I'm having the same problem now. By the way, I saw you made a Broden dataset with CIFAR-10, does it have semantic segmentation...

Don't know if it helps or not, but I created conda env with: conda install scipy==1.0.0 (must be installed first) conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch and Net-Dissection-Lite worked...

> @cyizhuo that worked perfectly, thank you so much! Glad to know that~

Don't know if it helps or not, but I created conda env with: conda install scipy==1.0.0 (must be installed first) conda install pytorch==1.2.0 torchvision==0.4.0 cudatoolkit=10.0 -c pytorch and Net-Dissection-Lite worked...

sloving ncnn bbox shifting problem: https://github.com/RangiLyu/nanodet/issues/456#issuecomment-1831356836

> Yes. You just need to hack some codes in `Tracking::GrabImageMonocular()` and overload `Frame` constructor with `GCNextractor`. I'm currently hacking Track.cc and Frame.cc, already spent half a day now, the...

> > > 是的。您只需要修改一些代码并`Tracking::GrabImageMonocular()`使用.`` Frame``GCNextractor `` > > > > > > 我目前正在破解Track.cc和Frame.cc，已经用了半天​​了，Tracking还是返回一个空的pose…… > > Did you succeed? Yes, I'll post my code later, but my result is that gcn...

> > > > > 是的。您只需要修改一些代码并`Tracking::GrabImageMonocular()`使用.` Frame`GCNextractor `` > > > > > > > > > > > > 我目前正在破解Track.cc和Frame.cc，已经用了半天​​了，Tracking还是返回一个空的pose…… > > > > > > > > >...