#define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:12:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
In file included from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:14:0:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:12:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
In file included from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:14:0:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:12:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/strings/string_view.h(495): warning: expression has no effect

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/platform/env.h(356): warning: overloaded virtual function "tensorflow::Env::RegisterFileSystem" is only partially overridden in class "tensorflow::EnvWrapper"

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=stream_executor::dnn::AlgorithmDesc]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(804): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=size_t]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(858): here

In file included from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:14:0:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_ops_types.cu:12:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/strings/string_view.h(495): warning: expression has no effect

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/platform/env.h(356): warning: overloaded virtual function "tensorflow::Env::RegisterFileSystem" is only partially overridden in class "tensorflow::EnvWrapper"

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=stream_executor::dnn::AlgorithmDesc]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(804): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=size_t]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(858): here

[ 18%] Building NVCC (Device) object src/cuda/CMakeFiles/cuda_gpu_backend.dir/cuda_gpu_backend_generated_approx_nn_conv_ops.cu.o
[ 27%] Building NVCC (Device) object src/cuda/CMakeFiles/cuda_gpu_backend.dir/cuda_gpu_backend_generated_approx_nn_conv_ops_gemm.cu.o
[ 36%] Building NVCC (Device) object src/cuda/CMakeFiles/cuda_gpu_backend.dir/cuda_gpu_backend_generated_approx_nn_conv_kernels.cu.o
In file included from /root/tf-approximate/tf2/src/cuda/../approx_nn_conv_kernels.h:18:0,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:14:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:13:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
In file included from /root/tf-approximate/tf2/src/cuda/../approx_nn_conv_kernels.h:18:0,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:14:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:13:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
In file included from /root/tf-approximate/tf2/src/cuda/../approx_nn_conv_kernels.h:18:0,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:14:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:13:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/strings/string_view.h(495): warning: expression has no effect

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/platform/env.h(356): warning: overloaded virtual function "tensorflow::Env::RegisterFileSystem" is only partially overridden in class "tensorflow::EnvWrapper"

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=stream_executor::dnn::AlgorithmDesc]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(804): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=size_t]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(858): here

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(254): warning: pointless comparison of unsigned integer with zero
          detected during instantiation of "void ApproxConvIm2ColKernel<GPUDevice, T, AT, NullApproxOpType_t>::operator()(const GPUDevice &, const NullApproxOpType_t<GPUDevice, T, AT> &, const T *, int, int, int, int, int, int, int, int, int, int, int, int, int, int, int, AT *, T *) [with T=Eigen::half, AT=Eigen::half]" 
(226): here

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(254): warning: pointless comparison of unsigned integer with zero
          detected during instantiation of "void ApproxConvIm2ColKernel<GPUDevice, T, AT, NullApproxOpType_t>::operator()(const GPUDevice &, const NullApproxOpType_t<GPUDevice, T, AT> &, const T *, int, int, int, int, int, int, int, int, int, int, int, int, int, int, int, AT *, T *) [with T=Eigen::half, AT=Eigen::half]" 
(226): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(701): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(706): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(711): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(726): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(735): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(737): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(760): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(769): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(771): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeededCommon(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
(823): here
            instantiation of "__nv_bool Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::evalSubExprsIfNeeded(Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::EvaluatorPointerType) [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(160): here
            instantiation of "__nv_bool Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::evalSubExprsIfNeeded(Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::EvaluatorPointerType) [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(629): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(829): warning: calling a __host__ function from a __host__ __device__ function is not allowed
          detected during:
            instantiation of "void Eigen::TensorReductionEvaluatorBase<const Eigen::TensorReductionOp<Op, Dims, ArgType, MakePointer_>, Device>::cleanup() [with Op=Eigen::internal::SumReducer<float>, Dims=const Eigen::array<int, 3UL>, ArgType=const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, MakePointer_=Eigen::MakePointer, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h(176): here
            instantiation of "void Eigen::TensorEvaluator<const Eigen::TensorAssignOp<LeftArgType, RightArgType>, Device>::cleanup() [with LeftArgType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, RightArgType=const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>, Device=Eigen::GpuDevice]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h(643): here
            instantiation of "void Eigen::internal::TensorExecutor<Expression, Eigen::GpuDevice, Vectorizable, Tiling>::run(const Expression &, const Eigen::GpuDevice &) [with Expression=const Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, const Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>>, Vectorizable=false, Tiling=Eigen::internal::Off]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h(35): here
            instantiation of "Eigen::TensorDevice<ExpressionType, DeviceType> &Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived &) [with ExpressionType=Eigen::TensorMap<Eigen::Tensor<float, 1, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, DeviceType=Eigen::GpuDevice, OtherDerived=Eigen::TensorReductionOp<Eigen::internal::SumReducer<float>, const Eigen::array<int, 3UL>, const Eigen::TensorMap<Eigen::Tensor<const float, 4, 1, Eigen::DenseIndex>, 16, Eigen::MakePointer>, Eigen::MakePointer>]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(302): here
            instantiation of "void ApproxFilterCorrCoeff<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, ConstTensor4<T>, Flat<T>) [with T=float, AT=tensorflow::uint8]" 
/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(305): here

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(608): warning: pointless comparison of unsigned integer with zero
          detected during instantiation of "void ApproxConvIm2ColKernel<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, const T *, int, int, int, int, int, int, int, int, int, int, int, int, int, int, int, AT *, T *) [with T=float, AT=tensorflow::uint8]" 
(627): here

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(608): warning: pointless comparison of unsigned integer with zero
          detected during instantiation of "void ApproxConvIm2ColKernel<Eigen::GpuDevice, T, AT, TableApproxOpType_t>::operator()(const Eigen::GpuDevice &, const TableApproxOpType_t<Eigen::GpuDevice, T, AT> &, const T *, int, int, int, int, int, int, int, int, int, int, int, int, int, int, int, AT *, T *) [with T=float, AT=tensorflow::uint8]" 
(627): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(706): warning: calling a __host__ function("Eigen::internal::FullReducer< ::Eigen::TensorReductionEvaluatorBase<const  ::Eigen::TensorReductionOp< ::Eigen::internal::SumReducer<float> , const  ::Eigen::array<int, (unsigned long)3ul> , const  ::Eigen::TensorMap< ::Eigen::Tensor<const float, (int)4, (int)1, long> , (int)16,  ::Eigen::MakePointer> ,  ::Eigen::MakePointer> ,  ::Eigen::GpuDevice> ,  ::Eigen::internal::SumReducer<float> ,  ::Eigen::GpuDevice, (bool)1> ::run<float> ") from a __host__ __device__ function("Eigen::TensorReductionEvaluatorBase<const  ::Eigen::TensorReductionOp< ::Eigen::internal::SumReducer<float> , const  ::Eigen::array<int, (unsigned long)3ul> , const  ::Eigen::TensorMap< ::Eigen::Tensor<const float, (int)4, (int)1, long> , (int)16,  ::Eigen::MakePointer> ,  ::Eigen::MakePointer> ,  ::Eigen::GpuDevice> ::evalSubExprsIfNeededCommon") is not allowed

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(735): warning: calling a __host__ function("Eigen::internal::InnerReducer< ::Eigen::TensorReductionEvaluatorBase<const  ::Eigen::TensorReductionOp< ::Eigen::internal::SumReducer<float> , const  ::Eigen::array<int, (unsigned long)3ul> , const  ::Eigen::TensorMap< ::Eigen::Tensor<const float, (int)4, (int)1, long> , (int)16,  ::Eigen::MakePointer> ,  ::Eigen::MakePointer> ,  ::Eigen::GpuDevice> ,  ::Eigen::internal::SumReducer<float> ,  ::Eigen::GpuDevice> ::run<float> ") from a __host__ __device__ function("Eigen::TensorReductionEvaluatorBase<const  ::Eigen::TensorReductionOp< ::Eigen::internal::SumReducer<float> , const  ::Eigen::array<int, (unsigned long)3ul> , const  ::Eigen::TensorMap< ::Eigen::Tensor<const float, (int)4, (int)1, long> , (int)16,  ::Eigen::MakePointer> ,  ::Eigen::MakePointer> ,  ::Eigen::GpuDevice> ::evalSubExprsIfNeededCommon") is not allowed

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/src/Tensor/TensorReduction.h(769): warning: calling a __host__ function("Eigen::internal::OuterReducer< ::Eigen::TensorReductionEvaluatorBase<const  ::Eigen::TensorReductionOp< ::Eigen::internal::SumReducer<float> , const  ::Eigen::array<int, (unsigned long)3ul> , const  ::Eigen::TensorMap< ::Eigen::Tensor<const float, (int)4, (int)1, long> , (int)16,  ::Eigen::MakePointer> ,  ::Eigen::MakePointer> ,  ::Eigen::GpuDevice> ,  ::Eigen::internal::SumReducer<float> ,  ::Eigen::GpuDevice> ::run") from a __host__ __device__ function("Eigen::TensorReductionEvaluatorBase<const  ::Eigen::TensorReductionOp< ::Eigen::internal::SumReducer<float> , const  ::Eigen::array<int, (unsigned long)3ul> , const  ::Eigen::TensorMap< ::Eigen::Tensor<const float, (int)4, (int)1, long> , (int)16,  ::Eigen::MakePointer> ,  ::Eigen::MakePointer> ,  ::Eigen::GpuDevice> ::evalSubExprsIfNeededCommon") is not allowed

In file included from /root/tf-approximate/tf2/src/cuda/../approx_nn_conv_kernels.h:18:0,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:14:
/root/tf-approximate/tf2/src/cuda/../approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_device_functions.h:32,
                 from /root/tf-approximate/tf2/src/cuda/../gpu_kernel_helper.h:25,
                 from /root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu:13:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/strings/string_view.h(495): warning: expression has no effect

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/platform/env.h(356): warning: overloaded virtual function "tensorflow::Env::RegisterFileSystem" is only partially overridden in class "tensorflow::EnvWrapper"

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=stream_executor::dnn::AlgorithmDesc]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(804): here

/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/absl/types/optional.h(425): warning: expression has no effect
          detected during instantiation of "const T &absl::optional<T>::operator*() const & [with T=size_t]" 
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/stream_executor/dnn.h(858): here

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(254): warning: pointless comparison of unsigned integer with zero

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(254): warning: pointless comparison of unsigned integer with zero

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(608): warning: pointless comparison of unsigned integer with zero

/root/tf-approximate/tf2/src/cuda/approx_nn_conv_kernels.cu(608): warning: pointless comparison of unsigned integer with zero

Scanning dependencies of target cuda_gpu_backend
[ 45%] Linking CXX static library libcuda_gpu_backend.a
[ 45%] Built target cuda_gpu_backend
Scanning dependencies of target ApproxGPUOpsTF
[ 54%] Building CXX object CMakeFiles/ApproxGPUOpsTF.dir/src/approx_nn_ops.cpp.o
[ 63%] Building CXX object CMakeFiles/ApproxGPUOpsTF.dir/src/approx_nn_conv_ops.cpp.o
In file included from /root/tf-approximate/tf2/src/approx_nn_conv_kernels.h:18:0,
                 from /root/tf-approximate/tf2/src/approx_nn_conv_ops.cpp:36:
/root/tf-approximate/tf2/src/approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/framework/numeric_types.h:20,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/framework/allocator.h:26,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/framework/op_kernel.h:25,
                 from /root/tf-approximate/tf2/src/approx_nn_conv_ops.cpp:23:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
[ 72%] Building CXX object CMakeFiles/ApproxGPUOpsTF.dir/src/approx_nn_conv_ops_gemm.cpp.o
[ 81%] Building CXX object CMakeFiles/ApproxGPUOpsTF.dir/src/approx_nn_conv_kernels.cpp.o
[ 90%] Building CXX object CMakeFiles/ApproxGPUOpsTF.dir/src/approx_nn_conv_ops_ref.cpp.o
In file included from /root/tf-approximate/tf2/src/approx_nn_conv_ops_ref.cpp:22:0:
/root/tf-approximate/tf2/src/approx_ops_types.h:18:0: warning: "EIGEN_STACK_ALLOCATION_LIMIT" redefined
 #define EIGEN_STACK_ALLOCATION_LIMIT 0
 
In file included from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/Core:19:0,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/Tensor:14,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/framework/numeric_types.h:20,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/framework/allocator.h:26,
                 from /usr/local/lib/python3.6/dist-packages/tensorflow_core/include/tensorflow/core/framework/op_kernel.h:25,
                 from /root/tf-approximate/tf2/src/approx_nn_conv_ops_ref.cpp:14:
/usr/local/lib/python3.6/dist-packages/tensorflow_core/include/unsupported/Eigen/CXX11/../../../Eigen/src/Core/util/Macros.h:54:0: note: this is the location of the previous definition
 #define EIGEN_STACK_ALLOCATION_LIMIT 131072
 
[100%] Linking CXX shared library libApproxGPUOpsTF.so
[100%] Built target ApproxGPUOpsTF