我跑了
cp Makefile.config.example Makefile.config
make all
按照网站上的建议完成安装。 我使用Ubuntu 14.04和CUDA以及OpenBlas。
错误消息显示如下
CXX / LD -o .build_release / tools / upgrade_net_proto_text.bin .build_release / lib / libcaffe.so:对caffe :: curandGetErrorString(curandStatus)的未定义引用
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: weight_gpu_gemm(double const *,double const *,double *)
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: forward_gpu_bias(double *,double const *)
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: forward_gpu_bias(float *,float const *)
.build_release / lib / libcaffe.so:对caffe :: cudnn :: dataType :: zero的未定义引用 .build_release / lib / libcaffe.so:对caffe :: cudnn :: dataType :: one的未定义引用 .build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: backward_gpu_gemm(float const *,float const *,float *)
.build_release / lib / libcaffe.so:对caffe :: cublasGetErrorString(cublasStatus_t)的未定义引用
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: forward_gpu_gemm(double const *,double const *,double *,bool)
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: backward_gpu_gemm(double const *,double const *,double *)
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: backward_gpu_bias(double *,double const *)
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: forward_gpu_gemm(float const *,float const *,float *,bool)
.build_release / lib / libcaffe.so:对caffe :: cudnn :: dataType :: zero的未定义引用 .build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: weight_gpu_gemm(float const *,float const *,float *)
.build_release / lib / libcaffe.so:对caffe的未定义引用:: BaseConvolutionLayer :: backward_gpu_bias(float *,float const *)
.build_release / lib / libcaffe.so:对caffe :: cudnn :: dataType :: one的未定义引用 collect2:错误:ld返回1退出状态
make:*** [.build_release / tools / upgrade_net_proto_text.bin]错误1
我只修改了Makefile.config。 修改后的Makefile.config如下所示
USE_CUDNN := 1
CUSTOM_CXX := g++
CUDA_DIR := /usr/local/cuda
-gencode arch=compute_20,code=sm_21 \
-gencode arch=compute_30,code=sm_30 \
-gencode arch=compute_35,code=sm_35 \
-gencode arch=compute_50,code=sm_50 \
-gencode arch=compute_50,code=compute_50
BLAS := OpenBlas
PYTHON_INCLUDE := /usr/include/python2.7 \
/usr/lib/python2.7/dist-packages/numpy/core/include
PYTHON_LIB := /usr/lib
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib
BUILD_DIR := build
DISTRIBUTE_DIR := distribute
TEST_GPUID := 0
Q ?= @
答案 0 :(得分:1)
您需要将BLAS
中的Makefile.config
设置更改为
BLAS := open
而不是'OpenBlas'
。