TRAIN_EVAL.md中的eval Bevformer指令有误

[niouoklul]

./tools/uniad_dist_train.sh ./projects/configs/bevformer/bevformer_base.py UniAD/ckpts/bevformer_r101_dcn_24ep.pth N_GPUS
应该改为
./tools/uniad_dist_eval.sh ./projects/configs/bevformer/base_bevformer.py ./ckpts/bevformer_r101_dcn_24ep.pth N_GPUS

[dd-xx-dot]

I'm getting an error during evaluation:
AssertionError: results must be a list from format_results in nuscenes_dataset.py.
I've checked the return value of forward_test, and it's already wrapping the result in a list, like [bbox_results].
However, the error still occurs.
Have you successfully resolved this issue before? If so, what did you do?

[>>>>>>>>>>>>>>>>>>>>>>>>>>] 6000/6000, 13.8 task/s, elapsed: 436s, ETA: 0s writing results to output/results.pkl Traceback (most recent call last): File "/root/workspace/UniAD/./tools/test.py", line 263, in <module> main() File "/root/workspace/UniAD/./tools/test.py", line 257, in main print(dataset.evaluate(outputs, **eval_kwargs)) File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/mmdet3d/datasets/nuscenes_dataset.py", line 503, in evaluate result_files, tmp_dir = self.format_results(results, jsonfile_prefix) File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/mmdet3d/datasets/nuscenes_dataset.py", line 443, in format_results assert isinstance(results, list), 'results must be a list' AssertionError: results must be a list ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 285776) of binary: /root/workspace/UniAD/venv/bin/python Traceback (most recent call last): File "/root/workspace/UniAD/venv/bin/torchrun", line 7, in <module> sys.exit(main()) File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/torch/distributed/elastic/multiprocessing/errors/__init__.py", line 346, in wrapper return f(*args, **kwargs) File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/torch/distributed/run.py", line 794, in main run(args) File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/torch/distributed/run.py", line 785, in run elastic_launch( File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/torch/distributed/launcher/api.py", line 134, in __call__ return launch_agent(self._config, self._entrypoint, list(args)) File "/root/workspace/UniAD/venv/lib/python3.9/site-packages/torch/distributed/launcher/api.py", line 250, in launch_agent raise ChildFailedError( torch.distributed.elastic.multiprocessing.errors.ChildFailedError: