IndexError: too many indices for tensor of dimension 1

[linrf123]

Traceback (most recent call last):
File "./tools/train.py", line 278, in
main()
File "./tools/train.py", line 267, in main
train_model(
File "/home/zhumf/linrf/FusionOcc-main/mmdet3d/apis/train.py", line 345, in train_model
train_detector(
File "/home/zhumf/linrf/FusionOcc-main/mmdet3d/apis/train.py", line 320, in train_detector
runner.run(data_loaders, cfg.workflow)
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/mmcv/runner/epoch_based_runner.py", line 130, in run
epoch_runner(data_loaders[i], **kwargs)
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/mmcv/runner/epoch_based_runner.py", line 51, in train
self.run_iter(data_batch, train_mode=True, **kwargs)
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/mmcv/runner/epoch_based_runner.py", line 29, in run_iter
outputs = self.model.train_step(data_batch, self.optimizer,
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/mmcv/parallel/distributed.py", line 59, in train_step
output = self.module.train_step(*inputs[0], **kwargs[0])
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/mmdet/models/detectors/base.py", line 248, in train_step
losses = self(**data)
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/mmcv/runner/fp16_utils.py", line 110, in new_func
return old_func(*args, **kwargs)
File "/home/zhumf/linrf/FusionOcc-main/mmdet3d/models/detectors/base.py", line 60, in forward
return self.forward_train(**kwargs)
File "/home/zhumf/linrf/FusionOcc-main/mmdet3d/models/detectors/fusion_occ.py", line 150, in forward_train
lidar_feat, x_list, x_sparse_out = self.lidar_encoder(points)
File "/home/zhumf/miniconda3/envs/fusionocc/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
return forward_call(*input, **kwargs)
File "/home/zhumf/linrf/FusionOcc-main/mmdet3d/models/backbones/lidar_encoder.py", line 195, in forward
batch_size = coords[-1, 0] + 1
IndexError: too many indices for tensor of dimension 1

Thank you for your work! I would like to ask why there is this error in the process of reproducing the code？

[ShuoZhang-code]

The input point cloud seems to be empty. Did the error occur during training or at the beginning? Did you make any modifications to the configuration?

[linrf123]

This happened late in the first round of training. I just changed the resolution of the input image down

[ShuoZhang-code]

I I didn't encounter this issue when I tested it locally. You can check if the input data is correct

[linrf123]

Thanks for your reply, may I ask you the version of each package in your environment @ShuoZhang-code

[ShuoZhang-code]

main prerequisites

Python = 3.8
nuscenes-devkit = 1.1.11
torch = 1.10.0
torch-scatter = 2.0.9
opencv-python = 4.9.0
Pillow = 10.0.1
mmcv-full = 1.5.3
mmsegmentation = 0.25.0
torchvision = 0.11.0
tensorboard = 2.14.0
mmdetection = 2.25.1