diff options
| author | cfRod <crefeda.rodrigues@arm.com> | 2021-11-05 11:29:53 +0000 |
|---|---|---|
| committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2021-11-09 10:19:03 +0000 |
| commit | f2c022e54ae65467cca4bc698b5b94e5b3c62c47 (patch) | |
| tree | dcf3f38e85ff9761573d2e7e37cc746291f828af /src/graph/nodes/FullyConnectedLayer.cpp | |
| parent | f349655cf75a686ed97e90d193a45602d6e82e6f (diff) | |
| download | ComputeLibrary-f2c022e54ae65467cca4bc698b5b94e5b3c62c47.tar.gz | |
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>
Diffstat (limited to 'src/graph/nodes/FullyConnectedLayer.cpp')
| -rw-r--r-- | src/graph/nodes/FullyConnectedLayer.cpp | 15 |
1 files changed, 12 insertions, 3 deletions
diff --git a/src/graph/nodes/FullyConnectedLayer.cpp b/src/graph/nodes/FullyConnectedLayer.cpp index 442f636b61..6278227878 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 arm_compute { 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) { |