sampepose / flownet2-tf

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks
MIT License
405 stars 195 forks source link

Error compiling cuda_kernel_helper.h: No such file or directory #100

Open kasrayazdani opened 5 years ago

kasrayazdani commented 5 years ago

I am not able to compile and was wondering if someone could help me. I am using tf 1.14, gcc 5.4, python 3.7, and cuda 10.1

Here is the make error message:

$ make nvcc --expt-relaxed-constexpr -g -std=c++11 -Ipython -c "import tensorflow; print(tensorflow.sysconfig.get_include())" -I"/usr/local/cuda-10.1/.." -DGOOGLE_CUDA=1 -DNDEBUG -D_MWAITXINTRIN_H_INCLUDED -D_FORCE_INLINES -D__STRICT_ANSI__ -D_GLIBCXX_USE_CXX11_ABI=0 -c src/ops/preprocessing/kernels/data_augmentation.cu.cc -x cu -Xcompiler -fPIC -o src/ops/build/data_augmentation.o src/ops/preprocessing/kernels/data_augmentation.cu.cc:7:53: fatal error: tensorflow/core/util/cuda_kernel_helper.h: No such file or directory compilation terminated. Makefile:62: recipe for target 'preprocessing' failed make: *** [preprocessing] Error 1

Apostatee commented 4 years ago

That would be tf version problem, there are tons of bugs and I gave up finally

wangzheng121 commented 2 years ago

我自己的亲身实践,tf1.13一切正常,只更改了tensorflow版本--1.15,然后就会报这个错误,我认为是tf版本的问题

candlecove-nju commented 2 years ago

Thank you! That really works! 谢谢!