An error occurs when test with PoinTr on my own dataset:subprocess.CalledProcessError

[zura-false]

Hi !
I am trying to train the PoinTr pretrained model on my own dataset use the method in #11 but after training to the last epoch I get these error :
` warnings.warn("Unable to load pointnet2_ops cpp extension. JIT Compiling.")
Traceback (most recent call last):
File "main.py", line 68, in
main()
File "main.py", line 64, in main
run_net(args, config, train_writer, val_writer)
File "/home/***/codes/PoinTr/tools/runner.py", line 162, in run_net
train_writer.close()
AttributeError: 'NoneType' object has no attribute 'close'

2022-04-27 12:12:53,925 - PoinTr - INFO - [Validation] EPOCH: 300 Metrics = ['0.3882', '14.8132', '0.7327']
2022-04-27 12:12:53,926 - PoinTr - INFO - ============================ TEST RESULTS ============================
2022-04-27 12:12:53,926 - PoinTr - INFO - Taxonomy #Sample F-Score CDL1 CDL2 #ModelName
2022-04-27 12:12:53,926 - PoinTr - INFO - 02691156 5 0.388 14.813 0.733 airplane
2022-04-27 12:12:53,927 - PoinTr - INFO - Overall 0.388 14.813 0.733
Killing subprocess 19184
Killing subprocess 19185
Traceback (most recent call last):
File "/home//anaconda3/envs/PoinTr/lib/python3.7/runpy.py", line 193, in _run_module_as_main
"main", mod_spec)
File "/home//anaconda3/envs/PoinTr/lib/python3.7/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/home//anaconda3/envs/PoinTr/lib/python3.7/site-packages/torch/distributed/launch.py", line 340, in
main()
File "/home//anaconda3/envs/PoinTr/lib/python3.7/site-packages/torch/distributed/launch.py", line 326, in main
sigkill_handler(signal.SIGTERM, None) # not coming back
File "/home//anaconda3/envs/PoinTr/lib/python3.7/site-packages/torch/distributed/launch.py", line 301, in sigkill_handler
raise subprocess.CalledProcessError(returncode=last_return_code, cmd=cmd)
subprocess.CalledProcessError: Command '['/home//anaconda3/envs/PoinTr/bin/python', '-u', 'main.py', '--local_rank=1', '--launcher', 'pytorch', '--sync_bn', '--config', './cfgs/JYG_models/PoinTr.yaml', '--exp_name', 'example']' returned non-zero exit status 1.
by running this code:CUDA_VISIBLE_DEVICES=0,1 bash ./scripts/dist_train.sh 2 13232
--config ./cfgs/MYDATA_models/PoinTr.yaml
--exp_name example`
I use two Nvidia 3060 GPUs
Could you please give some advice，thanks！

[yuxumin]

Hi,
see #32

[zura-false]

Hi, see #32

Thanks, this works