Run inference on single GPU

[kamalasubha]

Hi,
I am able to do all setup as per instructions given in README
In the evaluation step,

python -m torch.distributed.launch --nproc_per_node=4 tools/test.py --cfg config/lidar_rcnn.yaml --checkpoint outputs/lidar_rcnn/checkpoint_lidar_rcnn_59.pth.tar
python tools/create_results.py --cfg config/lidar_rcnn.yaml

I am facing the following questions while running the evaluation.

1. How to change the command to run a single GPU, nproc_per_node needs to be 1.
2. What should be MODEL.Frame number for checkpoint_lidar_rcnn_59.pth.tar?
   Since I am trying to understand the evaluation, kindly help me on this to fix.

[Lzc6996]

@kamalasubha

1. Note that, you should keep the nGPUS in config equal to nproc_per_node ， in your case, set both of them 1.
2. checkpoint_lidar_rcnn_59.pth.tar is trained by frame = 1

[kamalasubha]

@Lzc6996
Thanks for the inputs
I am facing following error with above config,

Traceback (most recent call last):
  File "tools/test.py", line 91, in <module>
    test(cfg, 0, valloader, model, device, cfg.TEST.TAT_PATH)
  File "/home/lidar/LiDAR_RCNN/src/LiDAR_RCNN/core/function.py", line 113, in test
    for idx, batch in enumerate(testloader):
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py", line 345, in __next__
    data = self._next_data()
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py", line 856, in _next_data
    return self._process_data(data)
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/dataloader.py", line 881, in _process_data
    data.reraise()
  File "/usr/local/lib/python3.6/dist-packages/torch/_utils.py", line 395, in reraise
    raise self.exc_type(msg)
ValueError: Caught ValueError in DataLoader worker process 0.
Original Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/worker.py", line 178, in _worker_loop
    data = fetcher.fetch(index)
  File "/usr/local/lib/python3.6/dist-packages/torch/utils/data/_utils/fetch.py", line 28, in fetch
    data.append(next(self.dataset_iter))
  File "/home/lidar/LiDAR_RCNN/src/LiDAR_RCNN/datasets/waymo/loader.py", line 87, in transform_test
    pcd_cur, pcd_pre, proposal, gt_box, gt_cls = load_data(it, self.frame)
  File "/home/lidar/LiDAR_RCNN/src/LiDAR_RCNN/datasets/waymo/data_utils.py", line 65, in load_data
    pcd_cur_ri1 = np.hstack([pcd_cur_ri1, pcd_add_ones_ri1])
  File "<__array_function__ internals>", line 6, in hstack
  File "/usr/local/lib/python3.6/dist-packages/numpy/core/shape_base.py", line 344, in hstack
    return _nx.concatenate(arrs, 0)
  File "<__array_function__ internals>", line 6, in concatenate
ValueError: all the input arrays must have same number of dimensions, but the array at index 0 has 1 dimension(s) and the array at index 1 has 2 dimension(s)

Traceback (most recent call last):
  File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main
    "__main__", mod_spec)
  File "/usr/lib/python3.6/runpy.py", line 85, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.6/dist-packages/torch/distributed/launch.py", line 263, in <module>
    main()
  File "/usr/local/lib/python3.6/dist-packages/torch/distributed/launch.py", line 259, in main
    cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python3', '-u', 'tools/test.py', '--local_rank=0', '--cfg', 'config/lidar_rcnn.yaml', '--checkpoint', '/home/lidar/models/checkpoint_lidar_rcnn_59.pth.tar']' returned non-zero exit status 1.

I used val tfrecords from allpreprocessor.zip that shared via mail. Any clue on this?

[Lzc6996]

@kamalasubha
I guess this because we update the code for multi-frame, but the val.tfrecord is generated by previous code.
Can you generate data by yourself following data_processer ?

[kamalasubha]

Ok @Lzc6996. I will generate val tf record based on the link given. But, can you please explain which part of the code needs to be changed to support a single frame? Does any separate demo exist for the same?

[Lzc6996]

@kamalasubha
I get another plan, you can use the old release version of our code. v0.1.1
Actually, I can't figure out why the two versions are incompatible without running it by myself.

[kamalasubha]

Thanks @Lzc6996 . I will look into it

[kamalasubha]

@Lzc6996 I am able to run with the older version. Thanks for the inputs