question about setup error

[gushengbo]

kaolin/csrc/ops/conversions/mesh_to_spc/mesh_to_spc_cuda.cu:21:10: fatal error: cub/device/device_scan.cuh: No such file or directory
#include <cub/device/device_scan.cuh>
^~~~~~~~~~~~~~~~~~~~~~~~~~~~

[Caenorst]

Hi @gushengbo please provide more information about your system / config and the full installation logs

[gushengbo]

Hi, I am the questioner of the previous question, now I successfully install kaolin==0.12.0 with ninja, but I get another bug when I run the code.

(icon) shengbo@gaia:~/ICON-master$ python -m apps.train -cfg ./configs/train/icon-filter.yaml -test
PyMeshLab 0.1.7 based on MeshLab 2020.12d
mesh............. True
ICON:
w/ Global Image Encoder: True
Image Features used by MLP: ['normal_F', 'normal_B']
Geometry Features used by MLP: ['sdf', 'cmap', 'norm', 'vis']
Dim of Image Features (local): 6
Dim of Geometry Features (ICON): 7
Dim of MLP's first layer: 13

GPU available: True, used: True
TPU available: None, using: 0 TPU cores
Resume MLP weights from ./data/ckpt/icon-filter.ckpt
Resume normal model from ./data/ckpt/normal.ckpt
load from ./data/cape/test.txt
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0,1,2,3]
cuda::::::: True
Testing: 0it [00:00, ?it/s]Traceback (most recent call last):
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/home/shengbo/ICON-master/apps/train.py", line 144, in
trainer.test(model=model, datamodule=datamodule)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 915, in test
results = self.__test_given_model(model, test_dataloaders)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 973, in __test_given_model
results = self.fit(model)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 499, in fit
self.dispatch()
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 540, in dispatch
self.accelerator.start_testing(self)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/accelerators/accelerator.py", line 76, in start_testing
self.training_type_plugin.start_testing(trainer)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py", line 118, in start_testing
self._results = trainer.run_test()
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 786, in run_test
eval_loop_results, _ = self.run_evaluation()
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/trainer.py", line 725, in run_evaluation
output = self.evaluation_loop.evaluation_step(batch, batch_idx, dataloader_idx)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/trainer/evaluation_loop.py", line 160, in evaluation_step
output = self.trainer.accelerator.test_step(args)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/accelerators/accelerator.py", line 195, in test_step
return self.training_type_plugin.test_step(*args)
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/pytorch_lightning/plugins/training_type/training_type_plugin.py", line 134, in test_step
return self.lightning_module.test_step(*args, **kwargs)
File "/home/shengbo/ICON-master/apps/ICON.py", line 572, in test_step
sdf = self.reconEngine(opt=self.cfg,
File "/home/shengbo/anaconda3/envs/icon/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1130, in _call_impl
return forward_call(*input, **kwargs)
File "/home/shengbo/ICON-master/lib/common/seg3d_lossless.py", line 147, in forward
return self._forward_faster(**kwargs)
File "/home/shengbo/ICON-master/lib/common/seg3d_lossless.py", line 169, in _forward_faster
occupancys = self.batch_eval(coords, **kwargs)
File "/home/shengbo/ICON-master/lib/common/seg3d_lossless.py", line 138, in batch_eval
occupancys = self.query_func(**kwargs, points=coords2D)
File "/home/shengbo/ICON-master/lib/common/train_util.py", line 434, in query_func
preds = netG.query(features=features,
File "/home/shengbo/ICON-master/lib/net/HGPIFuNet.py", line 307, in query
point_feat_out = point_feat_extractor.query(
File "/home/shengbo/ICON-master/lib/dataset/PointFeat.py", line 44, in query
residues, pts_ind, _ = point_to_mesh_distance(points, self.triangles)
File "/home/shengbo/ICON-master/kaolin/kaolin/metrics/trianglemesh.py", line 81, in point_to_mesh_distance
cur_dist, cur_face_idx, cur_dist_type = _UnbatchedTriangleDistanceCuda.apply(
File "/home/shengbo/ICON-master/kaolin/kaolin/metrics/trianglemesh.py", line 125, in forward
_C.metrics.unbatched_triangle_distance_forward_cuda(
RuntimeError: unbatched_triangle_distance not built with CUDA

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Wed_Oct_23_19:24:38_PDT_2019
Cuda compilation tools, release 10.2, V10.2.89

kaolin 0.12.0 /home/shengbo/ICON-master/kaolin
torch 1.12.1+cu102
torchaudio 0.12.1+cu102
torchmetrics 0.11.0
torchvision 0.13.1+cu102

I am in cluster.

[Caenorst]

Hi @gushengbo , I don't think you installed kaolin properly, can you please provide the installation logs?