[PyTorch] Official implementation of CVPR2022 paper "TransFusion: Robust LiDAR-Camera Fusion for 3D Object Detection with Transformers". https://arxiv.org/abs/2203.11496
I have searched related issues but cannot get the expected help.
Yes!
The issue has not been fixed in the latest version.
Using the latest versionDescribe the issue
_I'm currently trying to train NuScenes by running "python tools/train.py configs/transfusion_nusc_voxel_L.py". After fixing intial loading errors of the dataset, I now get a cuda error.
RuntimeError: CUDA error: no kernel image is available for execution on the device.
When I run a simple test using torch it uses the GPU just fine with current drivers and torch==1.5. What can I do to solve this? Specs listed under._
Environment
sys.platform: linux
Python: 3.7.12 | packaged by conda-forge | (default, Oct 26 2021, 06:08:21) [GCC 9.4.0]
CUDA available: True
GPU 0: NVIDIA GeForce RTX 4090
CUDA_HOME: /usr
NVCC: Build cuda_11.5.r11.5/compiler.30672275_0
GCC: gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
PyTorch: 1.5.0
PyTorch compiling details: PyTorch built with:
GCC 7.3
C++ Version: 201402
Intel(R) oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel(R) 64 architecture applications
Checklist
RuntimeError: CUDA error: no kernel image is available for execution on the device.
When I run a simple test using torch it uses the GPU just fine with current drivers and torch==1.5. What can I do to solve this? Specs listed under._
Environment sys.platform: linux Python: 3.7.12 | packaged by conda-forge | (default, Oct 26 2021, 06:08:21) [GCC 9.4.0] CUDA available: True GPU 0: NVIDIA GeForce RTX 4090 CUDA_HOME: /usr NVCC: Build cuda_11.5.r11.5/compiler.30672275_0 GCC: gcc (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0 PyTorch: 1.5.0 PyTorch compiling details: PyTorch built with:
TorchVision: 0.6.0a0+82fd1c8 OpenCV: 4.7.0 MMCV: 1.2.4 MMCV Compiler: GCC 7.5 MMCV CUDA Compiler: 10.1 MMDetection: 2.10.0 MMDetection3D: 0.11.0+unknown
Using CUDA 12.2, Cuda Toolkit release 11.5, Nvidia Geforce RTX 4090