liushuan

benchmark 使用GPU报错

3

## detail | 详细描述 | 詳細な説明 环境：vulkan + ncnn + NVIDIA2080TI + windows10 + VS2017 运行命令：benchncnn.exe 1 1 0 -1 时能正常运行， 运行命令：benchncnn.exe 1 1 0 0 时报错 compile_spirv_module failed -1...

发现 问题

1

dataset.py bbox_xywh = np.concatenate([(bbox_coor[2:] + bbox_coor[:2]) * 0.5, bbox_coor[2:] - bbox_coor[:2]], axis=-1) bbox_xywh_scaled = 1.0 * bbox_xywh[np.newaxis, :] / self.strides[:, np.newaxis] iou = [] exist_positive = False for i in...

deep_track_test: /home/liushuan/tools/eigen3/Eigen/src/Core/DenseStorage.h:109: Eigen::internal::plain_array::plain_array() [with T = float; int Size = 64; int MatrixOrArrayOptions = 1]: Assertion `(internal::UIntPtr(eigen_unaligned_array_assert_workaround_gcc47(array)) & (15)) == 0 && "this assertion is explained here: " "http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html" "...

nn::Sequential 的push_back 编译失败

1

nn::Sequential generator( // Layer 1 nn::Conv2d(nn::Conv2dOptions(3, 256, 4) .with_bias(false) .transposed(true)));

models 里面 的 bin和 param 不是用里面的 caffemodel 转的吗？ 为什么我用caffemodel 检测会出错，不准确呢？

1

models 里面 的 bin和 param 不是用里面的 caffemodel 转的吗？ 为什么我用caffemodel 检测会出错，不准确呢？

Eigen报内存对齐的错误

1

deep_track_test: /home/liushuan/tools/eigen3/Eigen/src/Core/DenseStorage.h:109: Eigen::internal::plain_array::plain_array() [with T = float; int Size = 64; int MatrixOrArrayOptions = 1]: Assertion `(internal::UIntPtr(eigen_unaligned_array_assert_workaround_gcc47(array)) & (15)) == 0 && "this assertion is explained here: " "http://eigen.tuxfamily.org/dox-devel/group__TopicUnalignedArrayAssert.html" "...