Enable fast_math in CpuFullyConnected
ONCPUML-529
* Add support for passing fast_math for fullyconnected layers via fc_info.
* Add support for passing fast_math to run ACL benchmark graphs.
* Add validation test and accuracy tests (updated fixtures).
Note: abs and rel. tolerance for fast math mode are set based on experimental data.
Signed-off-by: cfRod <crefeda.rodrigues@arm.com>
change-Id: Ib107d6264d3ae5e36555334f39a13e678f8618df
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/6521
Reviewed-by: SiCong Li <sicong.li@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/graph/nodes/FullyConnectedLayer.cpp b/src/graph/nodes/FullyConnectedLayer.cpp
index 442f636..6278227 100644
--- a/src/graph/nodes/FullyConnectedLayer.cpp
+++ b/src/graph/nodes/FullyConnectedLayer.cpp
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2018-2020 Arm Limited.
+ * Copyright (c) 2018-2021 Arm Limited.
*
* SPDX-License-Identifier: MIT
*
@@ -31,12 +31,21 @@
{
namespace graph
{
-FullyConnectedLayerNode::FullyConnectedLayerNode(unsigned int num_outputs, QuantizationInfo out_quant_info, FullyConnectedLayerInfo fc_info)
- : _num_outputs(num_outputs), _out_quant_info(std::move(out_quant_info)), _info(fc_info)
+FullyConnectedLayerNode::FullyConnectedLayerNode(unsigned int num_outputs, QuantizationInfo out_quant_info, FullyConnectedLayerInfo fc_info, FastMathHint fast_math_hint)
+ : _num_outputs(num_outputs), _out_quant_info(std::move(out_quant_info)), _info(fc_info), _fast_math_hint(fast_math_hint)
{
_input_edges.resize(3, EmptyEdgeID);
_outputs.resize(1, NullTensorID);
}
+void FullyConnectedLayerNode::set_fast_math_hint(FastMathHint hint)
+{
+ _fast_math_hint = hint;
+}
+
+FastMathHint FullyConnectedLayerNode::fast_math_hint() const
+{
+ return _fast_math_hint;
+}
void FullyConnectedLayerNode::set_fused_activation(ActivationLayerInfo fused_activation)
{