search

Banconxuan / RTM3D

The official PyTorch Implementation of RTM3D and KM3D for Monocular 3D Object Detection
MIT License
453 stars 85 forks source link

Error Running evaluate.py #31

Closed sparro12 closed 3 years ago

sparro12 commented 3 years ago

The following error shows up when running the evaluate.py script. I've included the command. It appears to be something with the iou file. The command earlier ran smoothly for the iou3d so it is not that. Looks like it would be not related to our setup. Thank you for any solutions in advance.

(KM3D) sparro12@instance-20210219-1359:~/KM3D/RTM3D$ python ./src/tools/kitti-object-eval-python/evaluate.py evaluate --label_path=/data/KITTI/training/label_2 --label_split_file /data/KITTI/ImageSets/val.txt --current_class=0,1,2 --coco=False --result_path=/data/KITTI/exp/results/data/ Traceback (most recent call last): File "./src/tools/kitti-object-eval-python/evaluate.py", line 4, in from eval import get_official_eval_result, get_coco_eval_result File "/home/sparro12/KM3D/RTM3D/src/tools/kitti-object-eval-python/eval.py", line 6, in from rotate_iou import rotate_iou_gpu_eval File "/home/sparro12/KM3D/RTM3D/src/tools/kitti-object-eval-python/rotate_iou.py", line 262, in def rotate_iou_kernel_eval(N, K, dev_boxes, dev_query_boxes, dev_iou, criterion=-1): File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/decorators.py", line 136, in kernel_jit targetoptions=targetoptions) File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/compiler.py", line 811, in init self.compile(sigs[0]) File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/compiler.py", line 952, in compile kernel.bind() File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/compiler.py", line 576, in bind self._func.get() File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/compiler.py", line 446, in get ptx = self.ptx.get() File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/compiler.py", line 416, in get self._extra_options) File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/cudadrv/nvvm.py", line 548, in llvm_to_ptx ptx = cu.compile(opts) File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/cudadrv/nvvm.py", line 236, in compile self._try_error(err, 'Failed to compile\n') File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/cudadrv/nvvm.py", line 254, in _try_error self.driver.check_error(err, "%s\n%s" % (msg, self.get_log())) File "/home/sparro12/miniconda3/envs/KM3D/lib/python3.6/site-packages/numba/cuda/cudadrv/nvvm.py", line 144, in check_error raise exc numba.cuda.cudadrv.error.NvvmError: Failed to compile

(91, 19): parse expected comma after load's type NVVM_ERROR_COMPILATION
Owen-Liuyuxuan commented 3 years ago

I met with the same phenomenon. For me the problem is that: Numba currently does not support CUDA 11.2 (not until 0.53.0)

I have been using docker to downgrade my CUDA version....

sparro12 commented 3 years ago

Thank you very much. We have multiple version of CUDA on the server. We will try to downgrade to 10.2.