Kyle Min
Could you explain more about this? I haven't resolved this problem yet..
Hi @Hrener Unfortunately, no.
I always recommend using the dataloader used in detectron2. You can refer to the dataloader of [this repo](https://github.com/MichiganCOG/Gaze-Attention/blob/main/dataset.py).
Hi, Thank you for your interest. We followed the standard convention of Deepmind/I3D. Please refer to the previous discussion: [link](https://github.com/kylemin/S3D/issues/4). All the training/testing details are in the original I3D paper.
Since TASED-Net has a series of max-pooling layers, it assumes that the spatial dimension of input (width and height) is a multiple of 16. I think you can resize your...
@iamhankai Thank you for your help. Now I have top1: 81.8%, top5: 95.9% for ViG-S. It looks pretty good (although they are not the same as the reported ones) @abhigoku10...
Hi Sagnik, We recently released a new repository named [GraVi-T](https://github.com/IntelLabs/GraVi-T). Please refer to it. Training time will be less than a few hours using a single gpu. Thank you, Kyle
Hi, I used 8 Titan Xp. I forgot the details because the experiments were done on my previous school server. Last time I discussed with the pytorch team, please see:...
Hi Arthur, I didn't have stability issues myself, but it totally makes sense that you have those issues. Over/under-flow of exp is a well-known issue so I think I once...
Hi Arthur, I am not sure how to address your problem and I don't recall how the output scores changed after training. I'll try to log in to my previous...