InceptText-Tensorflow
InceptText-Tensorflow copied to clipboard
FakerYFX
cd /lib;./make.sh error
/python2.7/site-packages/tensorflow/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/MathFunctions.h(1265): warning: calling a constexpr host function("real") from a host device function("abs") is not allowed. The experimental flag '--expt-relaxed-constexpr' can be used to allow this lib/python2.7/site-packages/tensorflow/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/MathFunctions.h(1270): warning: calling a constexpr host function("real") from a host device function("abs") is not allowed. The experimental flag '--expt-relaxed-constexpr' can be used to allow this. why?
中间有两个编译依赖错误，我都解决了，最后编译出来了，报了一个这个错误 tensorflow.python.framework.errors_impl.NotFoundError: /home/zhangtao/work/InceptText-Tensorflow/lib/deform_conv_layer/deform_conv.so: undefined symbol: _ZN10tensorflow8internal21CheckOpMessageBuilder9NewStringB5cxx11Ev
我这边是gcc5，是把下边注释打开，把上边注释掉是吧，Python是2.7
if you install tf using already-built binary, or gcc version 4.x, uncomment the two lines be low
g++ -std=c++11 -shared -o roi_pooling.so roi_pooling_op.cc
roi_pooling_op.cu.o -I $TF_INC -I $NSYNC_INC -fPIC -D GOOGLE_CUDA -lcudart -L $CUDA_HOM E/lib64 -L $TF_LIB -ltensorflow_framework -D_GLIBCXX_USE_CXX11_ABI=0
for gcc5-built tf
g++ -std=c++11 -shared -o roi_pooling.so roi_pooling_op.cc \
roi_pooling_op.cu.o -I $TF_INC -I $NSYNC_INC -fPIC -D GOOGLE_CUDA -lcudart -L $CUDA_HOM E/lib64 -L $TF_LIB -ltensorflow_framework -D_GLIBCXX_USE_CXX11_ABI=0
试了，把所有gcc5打开，上边的注释掉，还是这个错误tensorflow.python.framework.errors_impl.NotFoundError: /home/zhangtao/work/InceptText-Tensorflow/lib/deform_conv_layer/deform_conv.so: undefined symbol: _ZN10tensorflow8internal21CheckOpMessageBuilder9NewStringB5cxx11Ev
@zhangtao22 请问除了去gcc5相关掉注释，还用改其他值吗？我报错跟你不一样。
4 errors detected in the compilation of "/tmp/tmpxft_00005a43_000000007_deform_psroi_pooling_op_gpu.cu.cpp1.ii". g++: error: deform_psroi_pooling_op.cu.o: No such file or directory
undefined symbol: _ZTIN10tensorflow8OpKernelE**
我现在用虚拟机里面gcc4.9编译出来的deform_psroi_pooling.so替换了原来的同名库，还是有这个错误 undefined symbol: _ZTIN10tensorflow8OpKernelE
I also meet this error before
@xieyufei1993 Thank you for your advice . use the command pip install tensorflow-gpu==1.3 or compile tensorflow source code ?
我执行了pip install tensorflow-gpu==1.3.0 --user，然后进入lib文件夹下make，出现/usr/bin/ld: cannot find -ltensorflow_framework
@zhangtao22 设这个问题。各种我都试了 跑不起来 问题出在 “nvcc -std=c++11 -c -o deform_psroi_pooling_op.cu.o deform_psroi_pooling_op_gpu.cu.cc -I $TF_INC -D GOOGLE_CUDA=1 -x cu -Xcompiler -fPIC -D GOOGLE_CUDA -arch=$ARCH“” 不能生成deform_psroi_pooling_op.cu.o 弄了好多天 ，哎
@zhangtao22 好像跟g++版本没什么关系 tensorflow 1.0 1.3 1.4 1.11我都试了 都是can't find deform_psroi_pooling_op.cu.o 你这个问题 /usr/bin/ld: cannot find -ltensorflow_framework 1.4以上版本就没有了