sijie yan

@wuzieryun, @junhao1999, Thank you for your feedback. This bug has been fixed in the latest master branch.

@trungmanhhuynh, @yosagaf, This bug is caused by Pytorch 1.5.0. `nn.Parameter` and `nn.DataParallel` can not be used together. (Please refer to https://github.com/pytorch/pytorch/issues/36035 for more information) I recommend you to roll back...

Hi, if you use a smaller dataset, your model may suffer from over-fitting problem. May be you can use our model as a pre-trained model. Or, you can use some...

I think your problem is not caused by the limit of 2d pose. What is the accuracy? Sorry, I can not share the ntu-rgb dataset without their authorization. They replied...

It's strange. Did you check the value range of data and the definition of joints are identical?

Hi, @Zzoey, Thank you for your interests in our work. I roughly skim the dataset you mentioned. It seems to be much easier than Kinetics-skeleton dataset. [1] used a naive...

@Zzoey, can you share the config file and command you used? I think that will give me some hints.

hi @liuk3615, what labels you defined? what about the dataset size?

@Ai-is-light, @Hunkzer Hi, the old ST-GCN demo requires some heavy dependencies such as Openpose and Caffe. MMSkeleton has included other powerful pose estimators (e.g. HR-NET). Thus, we will release a...

Hi @yanqian123, I think our model is not so sensitive to the length of input sequences. Because of the different data distribution between two datasets, training from pretrained model is...