Some modifies to this package. Apply it and run python setup.py build_ext --inplace
in both thirdparty/visualDet3D/visualDet3D/networks/lib/ops/iou3d
and thirdparty/visualDet3D/visualDet3D/networks/lib/ops/dcn
-
thirdparty/visualDet3D/visualDet3D/networks/lib/ops/iou3d/src/iou3d_kernel.cpp
- 82: long * keep_data = keep.data_ptr(); -> int64_t * keep_data = keep.data_ptr<int64_t>();
- 100: unsigned long long remv_cpu[col_blocks]; -> unsigned long long* remv_cpu = new unsigned long long[col_blocks];
- 132: long * keep_data = keep.data_ptr(); -> int64_t * keep_data = keep.data_ptr<int64_t>();
- 150: unsigned long long remv_cpu[col_blocks]; -> unsigned long long* remv_cpu = new unsigned long long[col_blocks];
-
thirdparty/visualDet3D/visualDet3D/networks/lib/ops/iou3d/src/iou3d_kernel.cu
- 81+: const float EPS = 1e-8;
- 221+: const float EPS = 1e-8;
- 304+: const float EPS = 1e-8;
-
C:/repo/python/thirdparty/visualDet3D/visualDet3D/utils/utils.py
- 139+:
sys.path.insert(0, "config") cfg = getattr(importlib.import_module(os.path.splitext(cfg_filename)[0].split("/")[1]), 'cfg')
- Comment the rest
- 139+:
-
Installation pip3 install torch==1.11.0+cu113 torchvision==0.12.0+cu113 torchaudio===0.11.0+cu113 -f https://download.pytorch.org/whl/cu113/torch_stable.html pip3 install open3d tqdm opencv-python hydra-core scikit-image matplotlib plyfile pytorch_lightning argparse omegaconf --user