Add complex64 and complex128 gpu support for tensor_scatter_nd_add

This PR adds complex64 and complex128 gpu support for tensor_scatter_nd_add, as was raised in 40577. This PR fixes 40577. Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
2020-06-18 16:17:52 +00:00 · 2020-06-18 16:17:52 +00:00 · 9616883b82
commit 9616883b82
parent 06bc84b12e
1 changed files with 2 additions and 0 deletions
--- a/tensorflow/core/kernels/inplace_ops_functor_gpu.cu.cc
+++ b/tensorflow/core/kernels/inplace_ops_functor_gpu.cu.cc
@ -197,6 +197,8 @@ Status DoCopy(const Device& d, const Tensor& x, Tensor* y) {
    CASE(float)
    CASE(double)
    CASE(Eigen::half)
+    CASE(complex64)
+    CASE(complex128)
    CASE(int64)
 #undef CASE
    default: