Error occurs when running "initialSP_prepare_S3DIS.py"

[xuzijian-hik]

Hi! I met some errors when running "initialSP_prepare_S3DIS.py" with the command python data_prepare/initialSP_prepare_S3DIS.py.
Here is the output of the terminal：

xzj@xzj:~/githouse/GrowSP$ python data_prepare/initialSP_prepare_S3DIS.py
/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/site-packages/MinkowskiEngine/__init__.py:36: UserWarning: The environment variable `OMP_NUM_THREADS` not set. MinkowskiEngine will automatically set `OMP_NUM_THREADS=16`. If you want to set `OMP_NUM_THREADS` manually, please export it on the command line before running a python script. e.g. `export OMP_NUM_THREADS=12; python your_program.py`. It is recommended to set it below 24.
  warnings.warn(
start constructing initial superpoints
concurrent.futures.process._RemoteTraceback:
Traceback (most recent call last):
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/process.py", line 239, in _process_worker
    r = call_item.fn(*call_item.args, **call_item.kwargs)
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/process.py", line 198, in _process_chunk
    return [fn(*args) for args in chunk]
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/process.py", line 198, in <listcomp>
    return [fn(*args) for args in chunk]
  File "data_prepare/initialSP_prepare_S3DIS.py", line 78, in construct_superpoints
    coords = coords.numpy().astype(np.float32)
AttributeError: 'numpy.ndarray' object has no attribute 'numpy'
"""

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "data_prepare/initialSP_prepare_S3DIS.py", line 155, in <module>
    result = list(pool.map(construct_superpoints, path_list))
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/process.py", line 484, in _chain_from_iterable_of_lists
    for element in iterable:
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/_base.py", line 619, in result_iterator
    yield fs.pop().result()
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/_base.py", line 444, in result
    return self.__get_result()
  File "/home/xuzijian/anaconda3/envs/growsp2/lib/python3.8/concurrent/futures/_base.py", line 389, in __get_result
    raise self._exception
AttributeError: 'numpy.ndarray' object has no attribute 'numpy'

my enverinment is:

cudatookit=11.8
torch=2.0.0
torchaudio=2.0.1
torchvision=0.15.1
numpy=1.21.5
pandas=1.4.3
qhul=2015.2
pclpy=0.12.0
matplotlib=3.7.5
ninja=1.10.2.3
open3d=0.10.0.0
scikit-learn=0.22.2
scipy=1.10.1
seaborn=0.13.2

Although my CUDA version is 11.8, I think it may not be the reason of numpy error.
I also tried to uninstall numpy & pandas and then reinstall them, but the error still exists.
What can I do?

[zhang-zihui]

Hi, @xuzijian-hik,

It seems the error occurred in line78, this line is to covert the 'coords' from torch.tensor to numpy array. You can debug it to check if it is already a numpy array, and only convert its type to np.float32.

And I'm not sure your MinkowskiEngine version, is it 0.5.4?

Regards,
zihui

[xuzijian-hik]

Hi, @xuzijian-hik,

It seems the error occurred in line78, this line is to covert the 'coords' from torch.tensor to numpy array. You can debug it to check if it is already a numpy array, and only convert its type to np.float32.

And I'm not sure your MinkowskiEngine version, is it 0.5.4?

Regards, zihui

Thanks for your reply! I solve this problem by installing CUDA11.3. I think it might be something wrong with the MinkowskiEngine in CUDA11.8.

[MarcusnHuang]

Hello, I have also encountered the same problem. I installed CUDA11.3 and Minkowski Engine 0.5.4. May I ask if it is due to a version issue with Minkowski Engine? Looking forward to your reply

[zhang-zihui]

I have never encountered this issue in Mink0.5.4, so I'm not sure it. You can directly modify the code to convert output to numpy.

[zhang-zihui]

And CUDA 11.8 can also work for this code