haomo-ai / Cam4DOcc

[CVPR 2024] Cam4DOcc: Benchmark for Camera-Only 4D Occupancy Forecasting in Autonomous Driving Applications
MIT License
171 stars 9 forks source link

Error with running a baseline #5

Closed ekosman closed 3 months ago

ekosman commented 3 months ago

I experience the following problem when using config OCFNet_in_Cam4DOcc_V1.2.py:

  File "/home/eitan/Cam4DOcc/projects/occ_plugin/occupancy/image2bev/ViewTransformerLSSBEVDepth.py", line 495, in forward
    depth = self.depth_conv(depth)
  File "/fs/scratch/rb_bd_dlp_rng-dl01_cr_AID_employees/users/eitan/envs/cam4docc/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/fs/scratch/rb_bd_dlp_rng-dl01_cr_AID_employees/users/eitan/envs/cam4docc/lib/python3.7/site-packages/torch/nn/modules/container.py", line 141, in forward
    input = module(input)
  File "/fs/scratch/rb_bd_dlp_rng-dl01_cr_AID_employees/users/eitan/envs/cam4docc/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/fs/scratch/rb_bd_dlp_rng-dl01_cr_AID_employees/users/eitan/envs/cam4docc/lib/python3.7/site-packages/mmcv/ops/deform_conv.py", line 378, in forward
    False, self.im2col_step)
  File "/fs/scratch/rb_bd_dlp_rng-dl01_cr_AID_employees/users/eitan/envs/cam4docc/lib/python3.7/site-packages/mmcv/ops/deform_conv.py", line 109, in forward
    im2col_step=cur_im2col_step)
RuntimeError: CUDA error: no kernel image is available for execution on the device

What could be the problem? The installation seemed to work fine without any error.

BIT-MJY commented 3 months ago

@ekosman thanks for your interest in our work. This may be caused by the improper driver/package version. Please refer to the related repo such as this link for possible solutions.