使用cuDNN5编译py-faster-rcnn错误：cudnn.hpp(126): error: argument of type “int“ is incompatible ...

mjiansun

已于 2023-05-17 12:08:57 修改

阅读量1.4k

点赞数

分类专栏： Caffe 文章标签： caffe 人工智能深度学习

于 2019-05-07 15:43:46 首次发布

原文链接：https://blog.csdn.net/10km/article/details/62418583

版权

Caffe 专栏收录该内容

58 篇文章 5 订阅

订阅专栏

前几天在cuda8+cudnn5.1下编译py-faster-rcnn/caffe-fast-rcnn时报错了，

/home/guyadong/caffe/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp(126): error: argument of type “int” is incompatible with parameter of type “cudnnNanPropagation_t”

/home/guyadong/caffe/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp(126): error: too few arguments in function call

2 errors detected in the compilation of “/tmp/tmpxft_000067e2_00000000-5_rmsprop_solver.cpp4.ii”.
CMake Error at cuda_compile_generated_rmsprop_solver.cu.o.cmake:266 (message):
Error generating file
/home/guyadong/caffe/py-faster-rcnn/caffe-fast-rcnn/src/caffe/CMakeFiles/cuda_compile.dir/solvers/./cuda_compile_generated_rmsprop_solver.cu.o

src/caffe/CMakeFiles/caffe.dir/build.make:494: recipe for target ‘src/caffe/CMakeFiles/cuda_compile.dir/solvers/cuda_compile_generated_rmsprop_solver.cu.o’ failed
make2: * [src/caffe/CMakeFiles/cuda_compile.dir/solvers/cuda_compile_generated_rmsprop_solver.cu.o] Error 1
CMakeFiles/Makefile2:273: recipe for target ‘src/caffe/CMakeFiles/caffe.dir/all’ failed
make1: * [src/caffe/CMakeFiles/caffe.dir/all] Error 2
Makefile:116: recipe for target ‘all’ failed
make: * [all] Error 2

不得已改用cudnn4才算编译通过。

参见《cuda8+cudnn4 Faster R-CNN安装塈运行demo》

仔细研究，搞明白了原因：

py-faster-rcnn的作者在做这个项目时从bvlc/caffe(https://github.com/BVLC/caffe.git)项目创建了一个分支(那时还没有cudnn5),并在这个分支的基础上添加了自己的代码(主要是一些layer),形成现在的版本，就是py-faster-rcnn下的caffe-fast-rcnn。
所以py-faster-rcnn/caffe-fast-rcnn还是只支持cudnn4的代码，编译就会报错。

解决办法

在google上找了半天，总算有了办法：
其实BVLC/caffe目前已经支持cudnn5，要做的就是把BVLC/caffe的最新版本代码合并到py-faster-rcnn/caffe-fast-rcnn：

git版本合并

执行下面的命令实现版本合并

cd caffe-fast-rcnn
# 将BLVC/caffe添加为远程仓库，命名为caffe
git remote add caffe https://github.com/BVLC/caffe.git
# 获取BLVC/caffe代码
git fetch caffe
# 将BLVC/caffe的master分支合并到当前分支
# -X theirs 指定在合并遇到冲突时以BLVC/caffe版本为主
git merge -X theirs caffe/master

执行git merge -X theirs caffe/master会出现下面的界面要求输入注释信息：
不必输入任何内容，直接按Ctrl-X 离开

提示是否存储更动过的缓冲区，输入Y ,ENTER

要求写入的文件名:$MERGE_MSG,直接ENTER;

就完成了分支合并过程。

修改python_layer.hp

版本合并后，打开caffe-fast-rcnn/include/caffe/layers/python_layer.hp，self_.attr("phase") = static_cast<int>(this->phase_);这一行(line29)删除

然后再重新编译caffe-fast-rcnn,就可以正常编译通过了。

编译方式参见《cuda8+cudnn4 Faster R-CNN安装塈运行demo》

附：git输出

执行 git fetch caffe的输出

guyadong@gyd-u16:~/tmp/py-faster-rcnn/caffe-fast-rcnn$ git fetch caffe 
remote: Counting objects: 13860, done. 
remote: Compressing objects: 100% (8/8), done. 
remote: Total 13860 (delta 3378), reused 3374 (delta 3374), pack-reused 10478 
接收对象中: 100% (13860/13860), 13.76 MiB | 123.00 KiB/s, 完成. 
处理 delta 中: 100% (9800/9800), 完成 584 个本地对象. 
来自 https://github.com/BVLC/caffe 
* [新分支] device-abstraction -> caffe/device-abstraction 
* [新分支] gh-pages -> caffe/gh-pages 
* [新分支] master -> caffe/master 
* [新分支] opencl -> caffe/opencl 
* [新分支] parallel -> caffe/parallel 
* [新分支] tutorial -> caffe/tutorial 
* [新分支] williford-patch-1 -> caffe/williford-patch-1 
* [新分支] windows -> caffe/windows 
* [新标签] acm-mm-oss -> acm-mm-oss 
* [新标签] bsd -> bsd 
* [新标签] rc -> rc 
* [新标签] rc2 -> rc2 
* [新标签] rc3 -> rc3 
* [新标签] rcnn-release -> rcnn-release 
* [新标签] v0.1 -> v0.1 
* [新标签] v0.9 -> v0.9 
* [新标签] v0.99 -> v0.99 
* [新标签] v0.999 -> v0.999 
* [新标签] v0.9999 -> v0.9999 
* [新标签] rc4 -> rc4 
* [新标签] rc5 -> rc5

执行git merge -X theirs caffe/master的输出

自动合并 src/caffe/proto/caffe.proto 
自动合并 src/caffe/layers/dropout_layer.cpp 
删除 src/caffe/data_reader.cpp 
删除 scripts/travis/travis_setup_makefile_config.sh 
删除 scripts/travis/travis_install.sh 
删除 scripts/travis/travis_build_and_test.sh 
自动合并 python/caffe/_caffe.cpp 
自动合并 python/caffe/init.py 
自动合并 include/caffe/layers/python_layer.hpp 
自动合并 include/caffe/layer.hpp 
删除 include/caffe/data_reader.hpp 
删除 examples/hdf5_classification/solver.prototxt 
删除 examples/hdf5_classification/nonlinear_solver.prototxt 
删除 examples/03-fine-tuning.ipynb 
删除 examples/02-brewing-logreg.ipynb 
自动合并 LICENSE 
Merge made by the ‘recursive’ strategy. 
.github/ISSUE_TEMPLATE.md | 19 + 
.gitignore | 4 + 
.travis.yml | 58 +- 
CMakeLists.txt | 32 +- 
LICENSE | 15 +- 
Makefile | 40 +- 
Makefile.config.example | 15 +- 
cmake/ConfigGen.cmake | 67 +- 
cmake/Cuda.cmake | 35 +- 
cmake/Dependencies.cmake | 111 +- 
cmake/External/glog.cmake | 1 + 
cmake/Modules/FindAtlas.cmake | 6 +- 
cmake/Modules/FindMKL.cmake | 2 +- 
cmake/Modules/FindNCCL.cmake | 26 + 
cmake/Modules/FindvecLib.cmake | 7 +- 
cmake/ProtoBuf.cmake | 6 +- 
cmake/Summary.cmake | 1 + 
cmake/Targets.cmake | 27 +- 
cmake/Templates/CaffeConfig.cmake.in | 15 +- 
cmake/Templates/caffe_config.h.in | 19 - 
data/cifar10/get_cifar10.sh | 2 +- 
data/ilsvrc12/get_ilsvrc_aux.sh | 2 +- 
data/mnist/get_mnist.sh | 25 +- 
docker/README.md | 47 + 
docker/cpu/Dockerfile | 45 + 
docker/gpu/Dockerfile | 46 + 
docs/install_apt.md | 15 +- 
docs/install_apt_debian.md | 161 + 
docs/install_osx.md | 2 +- 
docs/install_yum.md | 4 +- 
docs/installation.md | 39 +- 
docs/multigpu.md | 2 +- 
docs/tutorial/layers.md | 562 +- 
docs/tutorial/layers/absval.md | 22 + 
docs/tutorial/layers/accuracy.md | 20 + 
docs/tutorial/layers/argmax.md | 18 + 
docs/tutorial/layers/batchnorm.md | 20 + 
docs/tutorial/layers/batchreindex.md | 16 + 
docs/tutorial/layers/bias.md | 19 + 
docs/tutorial/layers/bnll.md | 25 + 
docs/tutorial/layers/concat.md | 40 + 
docs/tutorial/layers/contrastiveloss.md | 20 + 
docs/tutorial/layers/convolution.md | 63 + 
docs/tutorial/layers/crop.md | 20 + 
docs/tutorial/layers/data.md | 29 + 
docs/tutorial/layers/deconvolution.md | 22 + 
docs/tutorial/layers/dropout.md | 20 + 
docs/tutorial/layers/dummydata.md | 20 + 
docs/tutorial/layers/eltwise.md | 20 + 
docs/tutorial/layers/elu.md | 25 + 
docs/tutorial/layers/embed.md | 20 + 
docs/tutorial/layers/euclideanloss.md | 16 + 
docs/tutorial/layers/exp.md | 24 + 
docs/tutorial/layers/filter.md | 15 + 
docs/tutorial/layers/flatten.md | 21 + 
docs/tutorial/layers/hdf5data.md | 20 + 
docs/tutorial/layers/hdf5output.md | 25 + 
docs/tutorial/layers/hingeloss.md | 19 + 
docs/tutorial/layers/im2col.md | 16 + 
docs/tutorial/layers/imagedata.md | 27 + 
docs/tutorial/layers/infogainloss.md | 23 + 
docs/tutorial/layers/innerproduct.md | 59 + 
docs/tutorial/layers/input.md | 19 + 
docs/tutorial/layers/log.md | 20 + 
docs/tutorial/layers/lrn.md | 28 + 
docs/tutorial/layers/lstm.md | 21 + 
docs/tutorial/layers/memorydata.md | 25 + 
docs/tutorial/layers/multinomiallogisticloss.md | 19 + 
docs/tutorial/layers/mvn.md | 20 + 
docs/tutorial/layers/parameter.md | 21 + 
docs/tutorial/layers/pooling.md | 47 + 
docs/tutorial/layers/power.md | 46 + 
docs/tutorial/layers/prelu.md | 20 + 
docs/tutorial/layers/python.md | 27 + 
docs/tutorial/layers/recurrent.md | 20 + 
docs/tutorial/layers/reduction.md | 20 + 
docs/tutorial/layers/relu.md | 32 + 
docs/tutorial/layers/reshape.md | 51 + 
docs/tutorial/layers/rnn.md | 19 + 
docs/tutorial/layers/scale.md | 20 + 
docs/tutorial/layers/sigmoid.md | 20 + 
docs/tutorial/layers/sigmoidcrossentropyloss.md | 13 + 
docs/tutorial/layers/silence.md | 17 + 
docs/tutorial/layers/slice.md | 42 + 
docs/tutorial/layers/softmax.md | 24 + 
docs/tutorial/layers/softmaxwithloss.md | 33 + 
docs/tutorial/layers/split.md | 17 + 
docs/tutorial/layers/spp.md | 20 + 
docs/tutorial/layers/tanh.md | 18 + 
docs/tutorial/layers/threshold.md | 18 + 
docs/tutorial/layers/tile.md | 20 + 
docs/tutorial/layers/windowdata.md | 19 + 
docs/tutorial/solver.md | 13 +- 
examples/00-classification.ipynb | 13031 +—————— 
examples/01-learning-lenet.ipynb | 4680 +—— 
examples/02-brewing-logreg.ipynb | 5771 ——– 
examples/02-fine-tuning.ipynb | 1175 ++ 
examples/03-fine-tuning.ipynb | 947 – 
examples/CMakeLists.txt | 2 +- 
examples/brewing-logreg.ipynb | 1164 ++ 
examples/cifar10/cifar10_full.prototxt | 11 +- 
…/cifar10/cifar10_full_sigmoid_solver.prototxt | 2 +- 
…/cifar10_full_sigmoid_solver_bn.prototxt | 2 +- 
examples/cifar10/cifar10_quick.prototxt | 11 +- 
examples/cifar10/convert_cifar_data.cpp | 2 + 
examples/cifar10/create_cifar10.sh | 1 + 
examples/cifar10/train_full.sh | 7 +- 
examples/cifar10/train_full_sigmoid.sh | 3 +- 
examples/cifar10/train_full_sigmoid_bn.sh | 3 +- 
examples/cifar10/train_quick.sh | 5 +- 
examples/cpp_classification/classification.cpp | 2 +- 
examples/cpp_classification/readme.md | 6 +- 
examples/finetune_flickr_style/readme.md | 10 +- 
…/hdf5_classification/nonlinear_solver.prototxt | 15 - 
examples/hdf5_classification/solver.prototxt | 15 - 
examples/imagenet/create_imagenet.sh | 1 + 
examples/imagenet/resume_training.sh | 4 +- 
examples/imagenet/train_caffenet.sh | 3 +- 
examples/images/cat gray.jpg | Bin 0 -> 92726 bytes 
examples/mnist/convert_mnist_data.cpp | 89 +- 
examples/mnist/create_mnist.sh | 1 + 
examples/mnist/lenet.prototxt | 11 +- 
examples/mnist/lenet_auto_solver.prototxt | 6 +- 
examples/mnist/readme.md | 2 +- 
examples/mnist/train_lenet.sh | 3 +- 
examples/mnist/train_lenet_adam.sh | 3 +- 
examples/mnist/train_lenet_consolidated.sh | 3 +- 
examples/mnist/train_lenet_docker.sh | 119 + 
examples/mnist/train_lenet_rmsprop.sh | 4 +- 
examples/mnist/train_mnist_autoencoder.sh | 3 +- 
examples/mnist/train_mnist_autoencoder_adadelta.sh | 3 +- 
examples/mnist/train_mnist_autoencoder_adagrad.sh | 3 +- 
examples/mnist/train_mnist_autoencoder_nesterov.sh | 3 +- 
examples/net_surgery.ipynb | 58 +- 
…/net_surgery/bvlc_caffenet_full_conv.prototxt | 15 +- 
examples/net_surgery/conv.prototxt | 11 +- 
examples/pascal-multilabel-with-datalayer.ipynb | 479 + 
…/pycaffe/layers/pascal_multilabel_datalayers.py | 216 + 
examples/pycaffe/tools.py | 121 + 
examples/siamese/create_mnist_siamese.sh | 1 + 
examples/siamese/mnist_siamese.prototxt | 13 +- 
examples/siamese/train_mnist_siamese.sh | 3 +- 
include/caffe/blob.hpp | 1 + 
include/caffe/common.hpp | 19 +- 
include/caffe/data_reader.hpp | 82 - 
include/caffe/internal_thread.hpp | 4 +- 
include/caffe/layer.hpp | 43 +- 
include/caffe/layer_factory.hpp | 2 +- 
include/caffe/layers/accuracy_layer.hpp | 2 +- 
include/caffe/layers/base_data_layer.hpp | 6 +- 
include/caffe/layers/batch_norm_layer.hpp | 29 +- 
include/caffe/layers/bias_layer.hpp | 10 +- 
include/caffe/layers/crop_layer.hpp | 76 + 
include/caffe/layers/cudnn_relu_layer.hpp | 1 + 
include/caffe/layers/cudnn_sigmoid_layer.hpp | 1 + 
include/caffe/layers/cudnn_tanh_layer.hpp | 1 + 
include/caffe/layers/data_layer.hpp | 7 +- 
include/caffe/layers/hdf5_data_layer.hpp | 6 +- 
include/caffe/layers/inner_product_layer.hpp | 1 + 
include/caffe/layers/input_layer.hpp | 44 + 
include/caffe/layers/lstm_layer.hpp | 154 + 
include/caffe/layers/parameter_layer.hpp | 45 + 
include/caffe/layers/python_layer.hpp | 5 +- 
include/caffe/layers/recurrent_layer.hpp | 187 + 
include/caffe/layers/rnn_layer.hpp | 47 + 
include/caffe/layers/scale_layer.hpp | 12 +- 
…/layers/sigmoid_cross_entropy_loss_layer.hpp | 18 + 
include/caffe/net.hpp | 65 +- 
include/caffe/parallel.hpp | 93 +- 
include/caffe/solver.hpp | 42 +- 
include/caffe/solver_factory.hpp | 2 +- 
include/caffe/syncedmem.hpp | 26 +- 
include/caffe/util/cudnn.hpp | 24 +- 
include/caffe/util/db_leveldb.hpp | 5 +- 
include/caffe/util/db_lmdb.hpp | 13 +- 
include/caffe/util/hdf5.hpp | 4 +- 
include/caffe/util/math_functions.hpp | 5 + 
include/caffe/util/mkl_alternate.hpp | 5 + 
include/caffe/util/nccl.hpp | 37 + 
include/caffe/util/upgrade_proto.hpp | 12 + 
matlab/+caffe/Net.m | 3 + 
matlab/+caffe/Solver.m | 3 + 
matlab/+caffe/private/caffe_.cpp | 26 +- 
matlab/CMakeLists.txt | 2 +- 
matlab/demo/classification_demo.m | 7 +- 
models/bvlc_alexnet/deploy.prototxt | 11 +- 
models/bvlc_googlenet/deploy.prototxt | 11 +- 
models/bvlc_googlenet/train_val.prototxt | 2 +- 
models/bvlc_reference_caffenet/deploy.prototxt | 11 +- 
…/bvlc_reference_rcnn_ilsvrc13/deploy.prototxt | 11 +- 
models/finetune_flickr_style/deploy.prototxt | 11 +- 
python/CMakeLists.txt | 28 +- 
python/caffe/init.py | 4 +- 
python/caffe/_caffe.cpp | 206 +- 
python/caffe/classifier.py | 1 + 
python/caffe/coord_map.py | 185 + 
python/caffe/detector.py | 2 +- 
python/caffe/draw.py | 40 +- 
python/caffe/io.py | 7 +- 
python/caffe/net_spec.py | 8 +- 
python/caffe/pycaffe.py | 107 +- 
python/caffe/test/test_coord_map.py | 192 + 
python/caffe/test/test_io.py | 15 + 
python/caffe/test/test_net.py | 273 +- 
python/caffe/test/test_python_layer.py | 26 + 
python/draw_net.py | 15 +- 
python/train.py | 100 + 
scripts/build_docs.sh | 3 + 
scripts/cpp_lint.py | 8 +- 
scripts/download_model_binary.py | 7 +- 
scripts/split_caffe_proto.py | 35 + 
scripts/travis/build.sh | 13 + 
scripts/travis/configure-cmake.sh | 32 + 
scripts/travis/configure-make.sh | 36 + 
scripts/travis/configure.sh | 11 + 
scripts/travis/defaults.sh | 10 + 
scripts/travis/install-deps.sh | 110 + 
scripts/travis/install-python-deps.sh | 14 + 
scripts/travis/setup-venv.sh | 18 + 
scripts/travis/test.sh | 19 + 
scripts/travis/travis_build_and_test.sh | 54 - 
scripts/travis/travis_install.sh | 99 - 
scripts/travis/travis_setup_makefile_config.sh | 31 - 
src/caffe/CMakeLists.txt | 16 +- 
src/caffe/blob.cpp | 22 +- 
src/caffe/common.cpp | 47 +- 
src/caffe/data_reader.cpp | 119 - 
src/caffe/data_transformer.cpp | 2 +- 
src/caffe/internal_thread.cpp | 10 +- 
src/caffe/layer.cpp | 20 - 
src/caffe/layer_factory.cpp | 7 + 
src/caffe/layers/base_data_layer.cpp | 44 +- 
src/caffe/layers/base_data_layer.cu | 21 +- 
src/caffe/layers/batch_norm_layer.cpp | 16 +- 
src/caffe/layers/crop_layer.cpp | 141 + 
src/caffe/layers/crop_layer.cu | 109 + 
src/caffe/layers/cudnn_conv_layer.cu | 12 +- 
src/caffe/layers/cudnn_relu_layer.cpp | 1 + 
src/caffe/layers/cudnn_relu_layer.cu | 23 +- 
src/caffe/layers/cudnn_sigmoid_layer.cpp | 2 + 
src/caffe/layers/cudnn_sigmoid_layer.cu | 23 +- 
src/caffe/layers/cudnn_tanh_layer.cpp | 1 + 
src/caffe/layers/cudnn_tanh_layer.cu | 23 +- 
src/caffe/layers/data_layer.cpp | 82 +- 
src/caffe/layers/dropout_layer.cpp | 4 +- 
src/caffe/layers/exp_layer.cpp | 3 +- 
src/caffe/layers/hdf5_data_layer.cpp | 62 +- 
src/caffe/layers/hdf5_data_layer.cu | 22 +- 
src/caffe/layers/image_data_layer.cpp | 24 +- 
src/caffe/layers/inner_product_layer.cpp | 42 +- 
src/caffe/layers/inner_product_layer.cu | 31 +- 
src/caffe/layers/input_layer.cpp | 27 + 
src/caffe/layers/loss_layer.cpp | 4 +- 
src/caffe/layers/lstm_layer.cpp | 244 + 
src/caffe/layers/lstm_unit_layer.cpp | 131 + 
src/caffe/layers/lstm_unit_layer.cu | 154 + 
src/caffe/layers/memory_data_layer.cpp | 2 +- 
src/caffe/layers/parameter_layer.cpp | 8 + 
src/caffe/layers/recurrent_layer.cpp | 295 + 
src/caffe/layers/recurrent_layer.cu | 44 + 
src/caffe/layers/rnn_layer.cpp | 236 + 
src/caffe/layers/scale_layer.cpp | 14 +- 
…/layers/sigmoid_cross_entropy_loss_layer.cpp | 79 +- 
…/layers/sigmoid_cross_entropy_loss_layer.cu | 80 +- 
src/caffe/layers/window_data_layer.cpp | 11 +- 
src/caffe/net.cpp | 171 +- 
src/caffe/parallel.cpp | 508 +- 
src/caffe/proto/caffe.proto | 101 +- 
src/caffe/solver.cpp | 53 +- 
src/caffe/solvers/adagrad_solver.cpp | 1 - 
src/caffe/solvers/nesterov_solver.cpp | 1 - 
src/caffe/solvers/sgd_solver.cpp | 4 +- 
src/caffe/syncedmem.cpp | 59 +- 
src/caffe/test/CMakeLists.txt | 2 +- 
src/caffe/test/test_blob.cpp | 8 + 
src/caffe/test/test_caffe_main.cpp | 3 - 
src/caffe/test/test_crop_layer.cpp | 283 + 
src/caffe/test/test_data_layer.cpp | 36 + 
src/caffe/test/test_data_transformer.cpp | 1 - 
src/caffe/test/test_embed_layer.cpp | 2 +- 
src/caffe/test/test_euclidean_loss_layer.cpp | 2 +- 
src/caffe/test/test_gradient_based_solver.cpp | 40 +- 
src/caffe/test/test_hdf5_output_layer.cpp | 10 +- 
src/caffe/test/test_hdf5data_layer.cpp | 32 +- 
src/caffe/test/test_image_data_layer.cpp | 44 +- 
src/caffe/test/test_inner_product_layer.cpp | 240 + 
src/caffe/test/test_lstm_layer.cpp | 288 + 
src/caffe/test/test_net.cpp | 308 +- 
src/caffe/test/test_neuron_layer.cpp | 20 + 
src/caffe/test/test_rnn_layer.cpp | 217 + 
…/test/test_sigmoid_cross_entropy_loss_layer.cpp | 28 + 
src/caffe/test/test_split_layer.cpp | 61 - 
src/caffe/util/benchmark.cpp | 3 +- 
src/caffe/util/blocking_queue.cpp | 5 - 
src/caffe/util/db_lmdb.cpp | 75 +- 
src/caffe/util/hdf5.cpp | 38 +- 
src/caffe/util/insert_splits.cpp | 21 +- 
src/caffe/util/math_functions.cu | 20 + 
src/caffe/util/upgrade_proto.cpp | 102 +- 
src/gtest/CMakeLists.txt | 3 + 
src/gtest/gtest-all.cpp | 4 +- 
src/gtest/gtest.h | 2 +- 
tools/caffe.cpp | 57 +- 
tools/convert_imageset.cpp | 9 +- 
tools/extra/extract_seconds.py | 8 + 
tools/extra/parse_log.py | 28 +- 
tools/extra/plot_log.gnuplot.example | 2 +- 
tools/extra/plot_training_log.py.example | 43 +- 
tools/extract_features.cpp | 5 +- 
tools/upgrade_net_proto_binary.cpp | 5 +- 
tools/upgrade_net_proto_text.cpp | 8 +- 
tools/upgrade_solver_proto_text.cpp | 3 +- 
312 files changed, 12807 insertions(+), 26312 deletions(-) 
create mode 100644 .github/ISSUE_TEMPLATE.md 
create mode 100644 cmake/Modules/FindNCCL.cmake 
create mode 100644 docker/README.md 
create mode 100644 docker/cpu/Dockerfile 
create mode 100644 docker/gpu/Dockerfile 
create mode 100644 docs/install_apt_debian.md 
create mode 100644 docs/tutorial/layers/absval.md 
create mode 100644 docs/tutorial/layers/accuracy.md 
create mode 100644 docs/tutorial/layers/argmax.md 
create mode 100644 docs/tutorial/layers/batchnorm.md 
create mode 100644 docs/tutorial/layers/batchreindex.md 
create mode 100644 docs/tutorial/layers/bias.md 
create mode 100644 docs/tutorial/layers/bnll.md 
create mode 100644 docs/tutorial/layers/concat.md 
create mode 100644 docs/tutorial/layers/contrastiveloss.md 
create mode 100644 docs/tutorial/layers/convolution.md 
create mode 100644 docs/tutorial/layers/crop.md 
create mode 100644 docs/tutorial/layers/data.md 
create mode 100644 docs/tutorial/layers/deconvolution.md 
create mode 100644 docs/tutorial/layers/dropout.md 
create mode 100644 docs/tutorial/layers/dummydata.md 
create mode 100644 docs/tutorial/layers/eltwise.md 
create mode 100644 docs/tutorial/layers/elu.md 
create mode 100644 docs/tutorial/layers/embed.md 
create mode 100644 docs/tutorial/layers/euclideanloss.md 
create mode 100644 docs/tutorial/layers/exp.md 
create mode 100644 docs/tutorial/layers/filter.md 
create mode 100644 docs/tutorial/layers/flatten.md 
create mode 100644 docs/tutorial/layers/hdf5data.md 
create mode 100644 docs/tutorial/layers/hdf5output.md 
create mode 100644 docs/tutorial/layers/hingeloss.md 
create mode 100644 docs/tutorial/layers/im2col.md 
create mode 100644 docs/tutorial/layers/imagedata.md 
create mode 100644 docs/tutorial/layers/infogainloss.md 
create mode 100644 docs/tutorial/layers/innerproduct.md 
create mode 100644 docs/tutorial/layers/input.md 
create mode 100644 docs/tutorial/layers/log.md 
create mode 100644 docs/tutorial/layers/lrn.md 
create mode 100644 docs/tutorial/layers/lstm.md 
create mode 100644 docs/tutorial/layers/memorydata.md 
create mode 100644 docs/tutorial/layers/multinomiallogisticloss.md 
create mode 100644 docs/tutorial/layers/mvn.md 
create mode 100644 docs/tutorial/layers/parameter.md 
create mode 100644 docs/tutorial/layers/pooling.md 
create mode 100644 docs/tutorial/layers/power.md 
create mode 100644 docs/tutorial/layers/prelu.md 
create mode 100644 docs/tutorial/layers/python.md 
create mode 100644 docs/tutorial/layers/recurrent.md 
create mode 100644 docs/tutorial/layers/reduction.md 
create mode 100644 docs/tutorial/layers/relu.md 
create mode 100644 docs/tutorial/layers/reshape.md 
create mode 100644 docs/tutorial/layers/rnn.md 
create mode 100644 docs/tutorial/layers/scale.md 
create mode 100644 docs/tutorial/layers/sigmoid.md 
create mode 100644 docs/tutorial/layers/sigmoidcrossentropyloss.md 
create mode 100644 docs/tutorial/layers/silence.md 
create mode 100644 docs/tutorial/layers/slice.md 
create mode 100644 docs/tutorial/layers/softmax.md 
create mode 100644 docs/tutorial/layers/softmaxwithloss.md 
create mode 100644 docs/tutorial/layers/split.md 
create mode 100644 docs/tutorial/layers/spp.md 
create mode 100644 docs/tutorial/layers/tanh.md 
create mode 100644 docs/tutorial/layers/threshold.md 
create mode 100644 docs/tutorial/layers/tile.md 
create mode 100644 docs/tutorial/layers/windowdata.md 
delete mode 100644 examples/02-brewing-logreg.ipynb 
create mode 100644 examples/02-fine-tuning.ipynb 
delete mode 100644 examples/03-fine-tuning.ipynb 
create mode 100644 examples/brewing-logreg.ipynb 
delete mode 100644 examples/hdf5_classification/nonlinear_solver.prototxt 
delete mode 100644 examples/hdf5_classification/solver.prototxt 
create mode 100644 examples/images/cat gray.jpg 
create mode 100755 examples/mnist/train_lenet_docker.sh 
create mode 100644 examples/pascal-multilabel-with-datalayer.ipynb 
create mode 100644 examples/pycaffe/layers/pascal_multilabel_datalayers.py 
create mode 100644 examples/pycaffe/tools.py 
delete mode 100644 include/caffe/data_reader.hpp 
create mode 100644 include/caffe/layers/crop_layer.hpp 
create mode 100644 include/caffe/layers/input_layer.hpp 
create mode 100644 include/caffe/layers/lstm_layer.hpp 
create mode 100644 include/caffe/layers/parameter_layer.hpp 
create mode 100644 include/caffe/layers/recurrent_layer.hpp 
create mode 100644 include/caffe/layers/rnn_layer.hpp 
create mode 100644 include/caffe/util/nccl.hpp 
mode change 100644 => 100755 models/bvlc_googlenet/train_val.prototxt 
create mode 100644 python/caffe/coord_map.py 
create mode 100644 python/caffe/test/test_coord_map.py 
create mode 100644 python/train.py 
create mode 100755 scripts/split_caffe_proto.py 
create mode 100755 scripts/travis/build.sh 
create mode 100644 scripts/travis/configure-cmake.sh 
create mode 100644 scripts/travis/configure-make.sh 
create mode 100755 scripts/travis/configure.sh 
create mode 100755 scripts/travis/defaults.sh 
create mode 100755 scripts/travis/install-deps.sh 
create mode 100755 scripts/travis/install-python-deps.sh 
create mode 100755 scripts/travis/setup-venv.sh 
create mode 100755 scripts/travis/test.sh 
delete mode 100755 scripts/travis/travis_build_and_test.sh 
delete mode 100755 scripts/travis/travis_install.sh 
delete mode 100755 scripts/travis/travis_setup_makefile_config.sh 
delete mode 100644 src/caffe/data_reader.cpp 
create mode 100644 src/caffe/layers/crop_layer.cpp 
create mode 100644 src/caffe/layers/crop_layer.cu 
create mode 100644 src/caffe/layers/input_layer.cpp 
create mode 100644 src/caffe/layers/lstm_layer.cpp 
create mode 100644 src/caffe/layers/lstm_unit_layer.cpp 
create mode 100644 src/caffe/layers/lstm_unit_layer.cu 
create mode 100644 src/caffe/layers/parameter_layer.cpp 
create mode 100644 src/caffe/layers/recurrent_layer.cpp 
create mode 100644 src/caffe/layers/recurrent_layer.cu 
create mode 100644 src/caffe/layers/rnn_layer.cpp 
create mode 100644 src/caffe/test/test_crop_layer.cpp 
create mode 100644 src/caffe/test/test_lstm_layer.cpp 
create mode 100644 src/caffe/test/test_rnn_layer.cpp

参考资料
《make error with cuda-7.5 and cundnn 5.1》
原文：https://blog.csdn.net/10km/article/details/62418583

mjiansun

关注

0
点赞
踩
0

收藏

觉得还不错? 一键收藏
1
评论
使用cuDNN5编译py-faster-rcnn错误：cudnn.hpp(126): error: argument of type “int“ is incompatible ...

前几天在cuda8+cudnn5.1下编译py-faster-rcnn/caffe-fast-rcnn时报错了，/home/guyadong/caffe/py-faster-rcnn/caffe-fast-rcnn/include/caffe/util/cudnn.hpp(126): error: argument of type “int” is incompatible with par...
复制链接

扫一扫

专栏目录