diff options
| author |  A. Unique TensorFlower <gardener@tensorflow.org> | 2017-09-19 09:09:05 -0700 |
|---|---|---|
| committer | TensorFlower Gardener <gardener@tensorflow.org> | 2017-09-19 09:12:47 -0700 |
| commit | 34018f8fa7290650291bbd478534e58c128a5df4 (patch) | |
| tree | d227f22d09089ee4cad773d490b39ecee4300018 /tensorflow/core/kernels/matrix_band_part_op.h | |
| parent | ec962ff63820e3ab9f5cc4c5f37c3579be0afcd9 (diff) | |
Add GPU support for QR decomposition.
Remove support support for on-the-fly transpose in internal matrix_band_part functor recently added (in anticipation of using it for QR), since it turned out to not be useful.
PiperOrigin-RevId: 169249336
Diffstat (limited to 'tensorflow/core/kernels/matrix_band_part_op.h')
| -rw-r--r-- | tensorflow/core/kernels/matrix_band_part_op.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tensorflow/core/kernels/matrix_band_part_op.h b/tensorflow/core/kernels/matrix_band_part_op.h index 43b6724dae..97cc950793 100644 --- a/tensorflow/core/kernels/matrix_band_part_op.h +++ b/tensorflow/core/kernels/matrix_band_part_op.h @@ -26,7 +26,7 @@ namespace functor { template <typename Device, typename Scalar> struct MatrixBandPartFunctor { void operator()(OpKernelContext* context, const Device& device, - int num_upper_diags, int num_lower_diags, bool transpose, + int num_upper_diags, int num_lower_diags, typename TTypes<Scalar, 3>::ConstTensor input, typename TTypes<Scalar, 3>::Tensor output); }; |