Commit 3bfd77c1 authored by Benjamin Kramer's avatar Benjamin Kramer Committed by TensorFlower Gardener
Browse files

[TF:XLA] Re-enable ResourceApplyAddSign of 16 bit floats 

This was blocked on an LLVM bug that was resolved in LLVM r336227. I had to increase the
fp tolerance to make the test pass.

 x: array([ 0.925781,  1.927734], dtype=float16)
 y: array([ 0.927734,  1.927734], dtype=float16)

PiperOrigin-RevId: 203808682
parent 826a19c3
Show whitespace changes
Inline Side-by-side
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Please to comment