Zzier 28d4edc647 fix(sconet): convert label_ids to CUDA LongTensor ...

- ScoNet:
  - Previously, `label_ids` remained a NumPy array, which could cause dtype/device mismatches when used with PyTorch tensors on GPU.
  - Convert `label_ids` to `torch.from_numpy(...).cuda().long()` to ensure correct tensor type (Long) and device (CUDA), aligning with loss functions that expect class indices on the same device.

2025-10-28 17:45:00 +08:00

..

backbones

Update ParsingGait (#160)

2023-10-29 21:53:02 +08:00

losses

remove dependency of pytorch_metric_learning

2023-10-09 18:32:23 +08:00

models

fix(sconet): convert label_ids to CUDA LongTensor

2025-10-28 17:45:00 +08:00

base_model.py

fix bug of evaluator_cfg

2024-03-20 18:10:12 +08:00

loss_aggregator.py

Reconstruct LossAggregator and fix some typos in config files (#100)

2023-01-14 17:33:38 +08:00

modules.py

Update modules.py

2025-08-14 23:01:51 +08:00