KNN_CUDA icon indicating copy to clipboard operation
KNN_CUDA copied to clipboard

pytorch knn [cuda version]

Results 10 KNN_CUDA issues
Sort by recently updated
recently updated
newest added

from knn_cuda import KNN File "E:\ProgramFiles\Anaconda3\envs\pytorch37\lib\site-packages\knn_cuda\__init__.py", line 38, in _knn = load_cpp_ext("knn") File "E:\ProgramFiles\Anaconda3\envs\pytorch37\lib\site-packages\knn_cuda\__init__.py", line 33, in load_cpp_ext with_cuda=True File "E:\ProgramFiles\Anaconda3\envs\pytorch37\lib\site-packages\torch\utils\cpp_extension.py", line 1091, in load keep_intermediates=keep_intermediates) File "E:\ProgramFiles\Anaconda3\envs\pytorch37\lib\site-packages\torch\utils\cpp_extension.py", line 1317,...

knn_cuda/csrc/_ext/knn/knn.so: cannot open shared object file: No such file or directory . how can I deal with this

when the points number is more than 50000, the OOM error appears. My GPU memory i s 10.92GiB, is there any way to reduce the memory occupy?

I've cloned the branch windows. After typing 'make' or 'python setup.py install', I got the message: ImportError: DLL load failed while importing knn OS: Windows 10 Python: 3.8 CUDA: 11.3...

D:\Anaconda\envs\torch1.8.2\lib\site-packages\torch\utils\cpp_extension.py:304: UserWarning: Error checking compiler version for cl: 'utf-8' codec can't decode byte 0xd3 in position 0: invalid continuation byte warnings.warn(f'Error checking compiler version for {compiler}: {error}') Traceback (most recent...

Hi, thanks for this nice code. I run the code on two machines with the same pytorch version but got garbage values on one of them. Here is the set...

Thanks for your great work! May I ask what do query and ref respectively stand for?

> File "/home/liy0r/anaconda3/envs/pt/lib/python3.7/site-packages/torch/utils/cpp_extension.py", line 1430, in verify_ninja_availability > raise RuntimeError("Ninja is required to load C++ extensions") > RuntimeError: Ninja is required to load C++ extensions I come across the problem...

Hi is this limited to only cuda 10.2?

I have conducted a test comparing the batch size of 1 with the code I implemented using the PyTorch API. The code is provided below: ```python import torch import time...