fast rcnn cudnn 5.0
2017.01.08 16:47
手动修改文件,参考了卜居大神的博客 http://blog.csdn.net/kkk584520/article/details/51163564方案1简单方便,但是当我们编译的是其他人修改过得源码,可能就会出错。方案2 步骤如下:
1. 用最新caffe源码的以下文件替换掉faster rcnn 的对应文件
include/caffe/layers/cudnn_relu_layer.hpp, src/caffe/layers/cudnn_relu_layer.cpp, src/caffe/layers/cudnn_relu_layer.cu
include/caffe/layers/cudnn_sigmoid_layer.hpp, src/caffe/layers/cudnn_sigmoid_layer.cpp, src/caffe/layers/cudnn_sigmoid_layer.cu
include/caffe/layers/cudnn_tanh_layer.hpp, src/caffe/layers/cudnn_tanh_layer.cpp, src/caffe/layers/cudnn_tanh_layer.cu
2. 用caffe源码中的这个文件替换掉faster rcnn 对应文件
include/caffe/util/cudnn.hpp
3. 将 faster rcnn 中的 src/caffe/layers/cudnn_conv_layer.cu 文件中的所有
cudnnConvolutionBackwardData_v3 函数名替换为 cudnnConvolutionBackwardData
cudnnConvolutionBackwardFilter_v3函数名替换为 cudnnConvolutionBackwardFilter
댓글 0
번호 | 제목 | 글쓴이 | 날짜 | 조회 수 |
---|---|---|---|---|
1624 | batch normalize | WHRIA | 2020.02.07 | 71 |
1623 | exif remove | WHRIA | 2020.01.28 | 69 |
1622 | nvidia docker | WHRIA | 2020.01.15 | 74 |
1621 | lvm 축소 확장 [4] | WHRIA | 2020.01.14 | 108 |
1620 | xml json pascal [4] | WHRIA | 2020.01.12 | 118 |
1619 | nvidia caffe | WHRIA | 2020.01.12 | 72 |
1618 | retinanet nvidia [3] | WHRIA | 2020.01.12 | 95 |
1617 | lvm [2] | WHRIA | 2020.01.09 | 75 |
1616 | usb 3.1 + DP | WHRIA | 2020.01.07 | 184 |
1615 | raid 6 rebuild | WHRIA | 2020.01.07 | 86 |
1614 | add extra raid disk | WHRIA | 2020.01.05 | 121 |
1613 | 참고 또 참고 | WHRIA | 2020.01.04 | 101 |
1612 | R graph | WHRIA | 2019.12.29 | 265 |
1611 | melanoma awareness [1] | WHRIA | 2019.12.27 | 74 |
1610 | 윈도우 raid ahci 전환 | WHRIA | 2019.12.26 | 92 |