commit 926f502ca731fa49bcdf949408ce25728616e5f2
author Murray Kornelsen <murray.kornelsen@mail.mcgill.ca> Wed Jul 13 21:22:39 2022 -0400
committer Pablo Marquez Tello <pablo.tello@arm.com> Wed Sep 14 09:15:03 2022 +0000
tree 7e221103a9c0c5c0e4c054abc07cbdf11c7c7b4e
parent 6e09e1404c635d948cf20eb6b4b5747dfb6656f2

Adding GELU activation

OpenCL implementation uses built in erf.

NEON implementation requires new vectorized erf.
Uses the following approximation:
erf(x) = 1 - 1 / (1 + a1x + a2x^2 + a3x^3 + a4x^4)^4
a1 = 0.278393, a2 = 0.230389, a3 = 0.000972, a4 = 0.078108

From https://en.wikipedia.org/wiki/Error_function#Numerical_approximations

Signed-off-by: Murray Kornelsen <murray.kornelsen@mail.mcgill.ca>
Change-Id: I2d3964b2c26a4334166b17135f9104bc6324fad2
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/7921
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Pablo Marquez Tello <pablo.tello@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

src/core/NEON/wrapper/intrinsics/intrinsics.h

diff --git a/src/core/NEON/wrapper/intrinsics/intrinsics.h b/src/core/NEON/wrapper/intrinsics/intrinsics.h
index 871d9cc..0256e0a 100644
--- a/src/core/NEON/wrapper/intrinsics/intrinsics.h
+++ b/src/core/NEON/wrapper/intrinsics/intrinsics.h
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2021 Arm Limited.
+ * Copyright (c) 2018-2022 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -39,6 +39,7 @@
 #include "src/core/NEON/wrapper/intrinsics/div.h"
 #include "src/core/NEON/wrapper/intrinsics/dup_n.h"
 #include "src/core/NEON/wrapper/intrinsics/eor.h"
+#include "src/core/NEON/wrapper/intrinsics/erf.h"
 #include "src/core/NEON/wrapper/intrinsics/exp.h"
 #include "src/core/NEON/wrapper/intrinsics/ext.h"
 #include "src/core/NEON/wrapper/intrinsics/gethigh.h"