runtime error

Container logs:

5h8bl 2023-03-14T16:13:14.457Z Traceback (most recent call last):
5h8bl 2023-03-14T16:13:14.457Z   File "app.py", line 14, in <module>
5h8bl 2023-03-14T16:13:14.457Z     local_rank, world_size = setup_model_parallel()
5h8bl 2023-03-14T16:13:14.457Z   File "/home/user/app/gen.py", line 25, in setup_model_parallel
5h8bl 2023-03-14T16:13:14.457Z     torch.distributed.init_process_group("nccl")
5h8bl 2023-03-14T16:13:14.457Z   File "/home/user/.local/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 761, in init_process_group
5h8bl 2023-03-14T16:13:14.457Z     default_pg = _new_process_group_helper(
5h8bl 2023-03-14T16:13:14.457Z   File "/home/user/.local/lib/python3.8/site-packages/torch/distributed/distributed_c10d.py", line 897, in _new_process_group_helper
5h8bl 2023-03-14T16:13:14.457Z     pg = ProcessGroupNCCL(prefix_store, group_rank, group_size, pg_options)
5h8bl 2023-03-14T16:13:14.457Z RuntimeError: ProcessGroupNCCL is only supported with GPUs, no GPUs found!