search

facebookresearch / maskrcnn-benchmark

Fast, modular reference implementation of Instance Segmentation and Object Detection algorithms in PyTorch.
MIT License
9.31k stars 2.49k forks source link

RuntimeError: cuda runtime error (48) : no kernel image is available for execution on the device at /maskrcnn_benchmark/csrc/cuda/ROIAlign_cuda.cu:297 #1314

Closed lqniunjunlper closed 3 years ago

lqniunjunlper commented 3 years ago

PyTorch version: 1.4.0+cu100 Is debug build: No CUDA used to build PyTorch: 10.0

OS: Ubuntu 16.04.6 LTS GCC version: (Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609 CMake version: Could not collect

Python version: 3.7 Is CUDA available: Yes CUDA runtime version: Could not collect GPU models and configuration: GPU 0: Tesla V100-SXM2-32GB Nvidia driver version: 450.80.02 cuDNN version: Could not collect

lqniunjunlper commented 3 years ago

GPU version NVIDIA-SMI 430.46 Driver Version: 440.64.00 CUDA Version: 10.2

Zhuobear commented 3 years ago

PyTorch 版本：1.4.0+cu100 是否調試構建：沒有 用於構建 PyTorch 的 CUDA：10.0

操作系統：Ubuntu 16.04.6 LTS GCC 版本：(Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609 CMake 版本：無法收集

Python 版本：3.7 CUDA 是否可用：是 CUDA 運行時版本：無法收集 GPU 模型和配置：GPU 0：Tesla V100-SXM2-32GB Nvidia 驅動程序版本：450.80.02 cuDNN 版本：無法收集

lqniunjunlper commented 3 years ago

fixed!

zhangyichent commented 2 years ago

How did you solve it？

MeinhardMark commented 1 year ago

PyTorch version: 1.4.0+cu100 Is debug build: No CUDA used to build PyTorch: 10.0

OS: Ubuntu 16.04.6 LTS GCC version: (Ubuntu 5.4.0-6ubuntu1~16.04.11) 5.4.0 20160609 CMake version: Could not collect

Python version: 3.7 Is CUDA available: Yes CUDA runtime version: Could not collect GPU models and configuration: GPU 0: Tesla V100-SXM2-32GB Nvidia driver version: 450.80.02 cuDNN version: Could not collect

How did you solve this?