lmbspecialops make error

[vivasvan1]

i am using
cuda version 9.2
python 3.6
tensorflow 1.12.0

i have used -D BUILD_WITH_CUDA=OFF argument during cmake which gives warning:

CMake Warning:
  Manually-specified variables were not used by the project:

    BUILD_WITH_CUDA

Make gives this error!

Scanning dependencies of target lmbspecialops
[ 55%] Building CXX object lib/CMakeFiles/lmbspecialops.dir/decode_flo_op.cc.o
/content/demon/lmbspecialops/src/decode_flo_op.cc: In member function ‘virtual void DecodeFloOp::Compute(tensorflow::OpKernelContext*)’:
/content/demon/lmbspecialops/src/decode_flo_op.cc:65:15: error: ‘const StringPiece {aka const class absl::string_view}’ has no member named ‘starts_with’
     if (!data.starts_with("PIEH")) {
               ^~~~~~~~~~~
lib/CMakeFiles/lmbspecialops.dir/build.make:118: recipe for target 'lib/CMakeFiles/lmbspecialops.dir/decode_flo_op.cc.o' failed
make[2]: *** [lib/CMakeFiles/lmbspecialops.dir/decode_flo_op.cc.o] Error 1
CMakeFiles/Makefile2:95: recipe for target 'lib/CMakeFiles/lmbspecialops.dir/all' failed
make[1]: *** [lib/CMakeFiles/lmbspecialops.dir/all] Error 2
Makefile:94: recipe for target 'all' failed
make: *** [all] Error 2

Can someone please explain what is this error?

[zhz218]

lmbspecialops should compile with tensorflow 1.4.0. or maybe refer to https://github.com/lmb-freiburg/lmbspecialops/tree/amiranas-tensorflow-1.9-patch for patch fixing of later tensorflow version.

[liutao0913]

change if (!data.starts_with("PIEH")) to if (data.substr(0, 4) != StringPiece("PIEH"))

[andylai2]

@liutao0913 's answer helped me fix the build errors. There are a few other places where the same substitution needs to be made.

[TheRustlessSummer]

@liutao0913 's answer helped me fix the build errors. There are a few other places where the same substitution needs to be made.

Hello, I am sorry to bother you. I have changed but there are still two errors like that:

In file included from /home/summer/anaconda3/lib/python3.6/site-packages/tensorflow/include/tensorflow/core/framework/attr_value_util.h:24:0,
from /home/summer/anaconda3/lib/python3.6/site-packages/tensorflow/include/tensorflow/core/framework/node_def_util.h:22,
from /home/summer/anaconda3/lib/python3.6/site-packages/tensorflow/include/tensorflow/core/framework/shape_inference.h:21,
from /home/summer/code/demon/lmbspecialops/src/decode_flo_op.cc:21:
/home/summer/anaconda3/lib/python3.6/site-packages/tensorflow/include/tensorflow/core/framework/tensor.h: In instantiation of ‘typename tensorflow::TTypes::ConstScalar tensorflow::Tensor::scalar() const [with T = std::basic_string; typename tensorflow::TTypes::ConstScalar = Eigen::TensorMap<Eigen::TensorFixedSize<const std::basic_string, Eigen::Sizes<>, 1, long int>, 16, Eigen::MakePointer>]’:
/home/summer/code/demon/lmbspecialops/src/decode_flo_op.cc:59:54: required from here
/home/summer/anaconda3/lib/python3.6/site-packages/tensorflow/include/tensorflow/core/framework/tensor.h:887:3: error: static assertion failed: std::string is no longer a scalar type, use tensorflow::tstring
static_assert(
^
lib/CMakeFiles/lmbspecialops.dir/build.make:6104: recipe for target 'lib/CMakeFiles/lmbspecialops.dir/decode_flo_op.cc.o' failed
make[2]: *** [lib/CMakeFiles/lmbspecialops.dir/decode_flo_op.cc.o] Error 1
CMakeFiles/Makefile2:127: recipe for target 'lib/CMakeFiles/lmbspecialops.dir/all' failed
make[1]: *** [lib/CMakeFiles/lmbspecialops.dir/all] Error 2
Makefile:94: recipe for target 'all' failed
make: *** [all] Error 2

[ghoshsuman]

@TheRustlessSummer Did you finally find a solution to your problem?