Could you upgrade the lib to support tensorflow 1.4, cudnn 6.0 and cuda 8.0?

[John1231983]

Hello, the original implementation has supported the cudnn 6.0 and cuda 8.0. Could you upgrade your lib to support the cudnn 6.0 and cuda 8.0. Because I am running in Tensorflow 1.4, so it does not support the lower version of cudnn (i.e 5.0 as your current code). Thanks for the great lib
P.s: I have close previous issue in other project due to wrong location

[Zardinality]

@John1231983 Sorry, I do not have time to do the upgrade right now. If you know where the problems locate, feel free to start a pull request.

[John1231983]

Thanks. The problem is that related to cuda.h. I checked in the tensorflow and noone can solve it in tensorflow 1.4. They only suggest downgrading to 1.2.1

/usr/local/lib/python3.5/dist-packages/tensorflow/include
nvcc warning : The 'compute_20', 'sm_20', and 'sm_21' architectures are deprecated, and may be removed in a future release (Use -Wno-deprecated-gpu-targets to suppress warning).
In file included from deform_conv.cu.cc:70:0:
/usr/local/lib/python3.5/dist-packages/tensorflow/include/tensorflow/core/util/cuda_kernel_helper.h:24:31: fatal error: cuda/include/cuda.h: No such file or directory
 #include "cuda/include/cuda.h"
                               ^

[John1231983]

Ok. Now I have succeeded to build in tensorflow 1.4 that install from pip3. This is my script, you can use it to update the README
nvcc file

TF_INC=$(python3 -c 'import tensorflow as tf; print(tf.sysconfig.get_include())')
echo $TF_INC
nvcc -std=c++11 -c -o deform_conv.cu.o deform_conv.cu.cc -I $TF_INC -D GOOGLE_CUDA=1 -x cu -Xcompiler -fPIC -L /usr/local/cuda-8.0/lib64/ -I /usr/local/ -I $TF_INC/external/nsync/public --expt-relaxed-constexpr -gencode arch=compute_52,code=sm_52

g++ file

TF_INC=$(python3 -c 'import tensorflow as tf; print(tf.sysconfig.get_include())')
TF_LIB=$(python -c 'import tensorflow as tf; print(tf.sysconfig.get_lib())')
echo $TF_INC
if [ ! -f $TF_INC/tensorflow/stream_executor/cuda/cuda_config.h ]; then
    cp ./cuda_config.h $TF_INC/tensorflow/stream_executor/cuda/
fi
g++ -std=c++11 -shared -o deform_conv.so deform_conv.cc deform_conv.cu.o -I $TF_INC -fPIC -lcudart -L $CUDA_HOME/lib64 -D GOOGLE_CUDA=1 -Wfatal-errors -I $CUDA_HOME/include -D_GLIBCXX_USE_CXX11_ABI=0 -L$TF_LIB -ltensorflow_framework -I $TF_INC/external/nsync/public

[Zardinality]

@John1231983 Thank you for the efforts! Unfortunately, I have no permission to install CuDNN 6 on the machine I use, so I can not cross-test whether these scripts work on my machine and make an update. But I update the README mentioning this problem and referring to this issue.