Building TF 1.13.1 from source


I see that NVIDIA provides us with Tensorflow v1.13.1 binary wheel installers, but can it be built from source? I keep running into errors in 1.13.1 (fine in r1.12). What are the steps for building from source in 1.13.1? Without building from source, examples like can’t be run.



Could you share the error you meet with us?
In general, you should be able to build TensorFlow with the similar steps of this topic:


Sure. I keep getting the following error:

Execution platform: @bazel_tools//platforms:host_platform
In file included from external/eigen_archive/unsupported/Eigen/CXX11/../../../Eigen/Core:296:0,
                 from external/eigen_archive/unsupported/Eigen/CXX11/Tensor:14,
                 from ./third_party/eigen3/unsupported/Eigen/CXX11/Tensor:1,
                 from ./tensorflow/contrib/rnn/kernels/lstm_ops.h:19,
                 from tensorflow/contrib/rnn/kernels/
external/eigen_archive/unsupported/Eigen/CXX11/../../../Eigen/src/Core/products/GeneralBlockPanelKernel.h: In member function 'void Eigen::internal::gebp_kernel<LhsScalar, RhsScalar, Index, DataMapper, mr, nr, ConjugateLhs, ConjugateRhs>::operator()(const DataMapper&, const LhsScalar*, const RhsScalar*, Index, Index, Index, Eigen::internal::gebp_kernel<LhsScalar, RhsScalar, Index, DataMapper, mr, nr, ConjugateLhs, ConjugateRhs>::ResScalar, Index, Index, Index, Index) [with LhsScalar = Eigen::half; RhsScalar = Eigen::half; Index = long int; DataMapper = Eigen::internal::blas_data_mapper<Eigen::half, long int, 0, 0>; int mr = 2; int nr = 4; bool ConjugateLhs = false; bool ConjugateRhs = false]':
external/eigen_archive/unsupported/Eigen/CXX11/../../../Eigen/src/Core/products/GeneralBlockPanelKernel.h:1879:3: internal compiler error: in emit_move_insn, at expr.c:3698
Please submit a full bug report,
with preprocessed source if appropriate.

I’m using bazel 0.21.0.


This is a TensorFlow issue with gcc7.3.1:

It’s recommended to check with TensorFlower for a possible workaround first.