Zelin Zhao

I understand your issues and I think installing DCN v2 via pip is not possible currently. This is because the prebuilt dcn via pip is not compatible with your local...

It should be deformable cnn, not deformable detr. Sorry foe the confusion.

You may not find the exact issue but the problem here is clear that your lib versions are not compatible so you cannot import dcn v2. https://www.google.com/amp/s/varhowto.com/check-pytorch-cuda-version/%3Famp

I didn't meet the problems before. You can first look at these issues: https://github.com/fundamentalvision/Deformable-DETR/issues?q=is%3Aissue+cuda+is+not+available+is%3Aclosed https://github.com/fundamentalvision/Deformable-DETR/issues/112 https://github.com/fundamentalvision/Deformable-DETR/issues/88 It seems that you cannot find CUDA library on your server. You can check...

No, seemingly this cannot be used. Try to save per frame images one by one and generate video from them.

> Hey guys，Have you ever recorded agents gameplay and save it as a video？ @sjtuytc @ryanprinster sure, what's your problem

> Is the flags(such as --demofiles, --record, --video, and --demo) useful？Now I'm reproducing the Impala algorithm of [deepmind,](https://github.com/deepmind/scalable_agent) but I don't know how to visualize it. I don't know if...

Sure, write it by your self.

I guess the easiest thing is to consider use a workaround. For example, 1) create another environment and perform actions ahead. 2) adjust your algorithm so that only previous state...

you need to go to the source code and modify the DEBUG callback.