VISTA icon indicating copy to clipboard operation
VISTA copied to clipboard

How can I train on nuscenes v1.0-mini?

Open zheng0819 opened this issue 1 year ago • 1 comments

The memory of my computer is not enough to train on nuscenes v1.0, so I try to train on nuscenes v1.0-mini. I change the config file, but an error occurred while training : Caught ValueError in DataLoader worker process 0. Original Traceback (most recent call last): File "/home/yusen0819/anaconda3/envs/vista/lib/python3.7/site-packages/torch/utils/data/_utils/worker.py", line 202, in _worker_loop data = fetcher.fetch(index) File "/home/yusen0819/anaconda3/envs/vista/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in fetch data = [self.dataset[idx] for idx in possibly_batched_index] File "/home/yusen0819/anaconda3/envs/vista/lib/python3.7/site-packages/torch/utils/data/_utils/fetch.py", line 44, in data = [self.dataset[idx] for idx in possibly_batched_index] File "/home/yusen0819/VISTA/det3d/datasets/nuscenes/nuscenes.py", line 193, in getitem return self.get_sensor_data(idx) File "/home/yusen0819/VISTA/det3d/datasets/nuscenes/nuscenes.py", line 189, in get_sensor_data data, _ = self.pipeline(res, info) File "/home/yusen0819/VISTA/det3d/datasets/pipelines/compose.py", line 25, in call res, info = t(res, info) File "/home/yusen0819/VISTA/det3d/datasets/pipelines/preprocess.py", line 96, in call gt_drop_max_keep=self.gt_drop_max_keep, File "/home/yusen0819/VISTA/det3d/core/sampler/sample_ops.py", line 247, in sample_all "points": np.concatenate(s_points_list, axis=0), File "<array_function internals>", line 6, in concatenate ValueError: need at least one array to concatenate File "/home/yusen0819/VISTA/det3d/torchie/trainer/trainer.py", line 389, in train for i, data_batch in enumerate(data_loader): File "/home/yusen0819/VISTA/det3d/torchie/trainer/trainer.py", line 542, in run epoch_runner(data_loaders[i], self.epoch, **kwargs) File "/home/yusen0819/VISTA/det3d/torchie/apis/train.py", line 336, in train_detector trainer.run(data_loaders, cfg.workflow, cfg.total_epochs, local_rank=cfg.local_rank) File "/home/yusen0819/VISTA/tools/train.py", line 126, in main logger=logger, File "/home/yusen0819/VISTA/tools/train.py", line 131, in main() I cannot handle this error. How can I train on nuscenes v1.0-mini? Thanks.

zheng0819 avatar Mar 24 '23 00:03 zheng0819