Jinkai Zheng

Hi~ Because the framework is based on DDP mode, it is recommended that you use only 1 GPU for debugging. This will make it clear for you to examine your...

Just like this, change the value of **CUDA_VISIBLE_DEVICES** and **--nproc_per_node**: `CUDA_VISIBLE_DEVICES=0 python -m torch.distributed.launch --nproc_per_node=1 lib/main.py --cfgs ./config/smplgait_64pixel.yaml --phase train`

I recommend you start at the [source](https://github.com/Gait3D/Gait3D-Benchmark/blob/72beab994c137b902d826f4b9f9e95b107bebd78/lib/data/datasets/dataset_smplgait.py#L87) and go step by step to make sure what is causing the missing of smpl data.

The "enable_float16" in trainer_cfg aims to memory reduction and speed up. Maybe you can try: `smpls = [np2var(np.asarray([fra for fra in smpl]).astype(float), requires_grad=requires_grad).float() for smpl in smpls_batch]`

Hi~ Because [the GREW's official evaluation website](https://codalab.lisn.upsaclay.fr/competitions/3409) was not available when we worked on Gait3D. To achieve the cross-domain experiment, we randomly sample 1,000 IDs from the training set for...