raise ValueError("batch_size should be a positive integer value, " ValueError: batch_size should be a positive integer value, but got batch_size=0

[LTyanghuang]

What's the feature?

2024-09-19 02:42:59,779 - mmcv - INFO -
bbox_head.retina_cls.weight - torch.Size([135, 256, 3, 3]):
NormalInit: mean=0, std=0.01, bias=-4.59511985013459

2024-09-19 02:42:59,779 - mmcv - INFO -
bbox_head.retina_cls.bias - torch.Size([135]):
NormalInit: mean=0, std=0.01, bias=-4.59511985013459

2024-09-19 02:42:59,779 - mmcv - INFO -
bbox_head.retina_reg.weight - torch.Size([45, 256, 3, 3]):
NormalInit: mean=0, std=0.01, bias=0

2024-09-19 02:42:59,779 - mmcv - INFO -
bbox_head.retina_reg.bias - torch.Size([45]):
NormalInit: mean=0, std=0.01, bias=0

Traceback (most recent call last):
File "tools/train.py", line 194, in
main()
File "tools/train.py", line 183, in main
train_detector(
File "/root/dota_workspace/mmrotate/mmrotate/apis/train.py", line 83, in train_detector
data_loaders = [build_dataloader(ds, **train_loader_cfg) for ds in dataset]
File "/root/dota_workspace/mmrotate/mmrotate/apis/train.py", line 83, in
data_loaders = [build_dataloader(ds, **train_loader_cfg) for ds in dataset]
File "/local/miniconda/envs/moca_cv/lib/python3.8/site-packages/mmdet/datasets/builder.py", line 195, in build_dataloader
data_loader = DataLoader(
File "/local/miniconda/envs/moca_cv/lib/python3.8/site-packages/torch/utils/data/dataloader.py", line 274, in init
batch_sampler = BatchSampler(sampler, batch_size, drop_last)
File "/local/miniconda/envs/moca_cv/lib/python3.8/site-packages/torch/utils/data/sampler.py", line 218, in init
raise ValueError("batch_size should be a positive integer value, "
ValueError: batch_size should be a positive integer value, but got batch_size=0

Any other context?

在训练的时候报错bacth_size=0;
python tools/train.py configs/rotated_retinanet/rotated_retinanet_obb_r50_fpn_1x_dota_ms_rr_le90.py --resume-from rotated_retinanet_obb_r50_fpn_1x_dota_ms_rr_le90-1da1ec9c.pth --work-dir logs/ --gpus 0