Wang-Bo-xuan
Wang-Bo-xuan
没有找到触发规律,大多数是开机自启程序时发生,后续使用很少出现 sys info: 3840 x 2160 16:9的屏幕 为了适应部分低dpi程序,系统设置里设置了200%的ui拉伸 系统ubuntu 18.04 bionic
 
`Linking CXX executable /home/wangboxuan/Desktop/map_ws/devel/lib/interactive_slam/interactive_slam CMakeFiles/interactive_slam.dir/src/hdl_graph_slam/plane_detection_window.cpp.o:(.data.rel.ro._ZTVN3pcl25SampleConsensusModelPlaneINS_9PointXYZIEEE[_ZTVN3pcl25SampleConsensusModelPlaneINS_9PointXYZIEEE]+0x28): undefined reference to `pcl::SampleConsensusModelPlane::computeModelCoefficients(std::vector const&, Eigen::Matrix&) const' CMakeFiles/interactive_slam.dir/src/hdl_graph_slam/plane_detection_window.cpp.o:(.data.rel.ro._ZTVN3pcl25SampleConsensusModelPlaneINS_9PointXYZIEEE[_ZTVN3pcl25SampleConsensusModelPlaneINS_9PointXYZIEEE]+0x30): undefined reference to `pcl::SampleConsensusModelPlane::optimizeModelCoefficients(std::vector const&, Eigen::Matrix const&, Eigen::Matrix&) const' CMakeFiles/interactive_slam.dir/src/hdl_graph_slam/plane_detection_window.cpp.o:(.data.rel.ro._ZTVN3pcl25SampleConsensusModelPlaneINS_9PointXYZIEEE[_ZTVN3pcl25SampleConsensusModelPlaneINS_9PointXYZIEEE]+0x38): undefined reference to `pcl::SampleConsensusModelPlane::getDistancesToModel(Eigen::Matrix const&, std::vector&) const'...
首先感谢您优秀的工作! 我在一台windows11上尝试训练,配置为32g + 2080s; cuda版本: > | NVIDIA-SMI 572.47 Driver Version: 572.47 CUDA Version: 12.8 | nvcc版本: > Cuda compilation tools, release 12.8, V12.8.61 cudnn版本: > v9.7 使用anaconda维护虚拟环境;之前有尝试过几个数值回归的神经网络demo,安装的torch version=2.5.1;是可以调用gpu的; 在尝试本工程时,使用了requ文件,可以正常启动模型及对话,目前也可以进行预训练,但是无法调用gpu;想了解下如何解决?...