Add an option to use lowest for max-pooling
- Add a parameter in PoolingLayerInfo class to pick which value to use as min for max-pooling.
Resolves: [ONCPUML-1166]
Signed-off-by: Adnan AlSinan <adnan.alsinan@arm.com>
Change-Id: I34e1cccc15176bbf31523c61e99f3188ddca23e1
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8989
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
Reviewed-by: Pablo Marquez Tello <pablo.tello@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/gpu/cl/kernels/ClPool2dKernel.cpp b/src/gpu/cl/kernels/ClPool2dKernel.cpp
index 2c98c59..83bc6bb 100644
--- a/src/gpu/cl/kernels/ClPool2dKernel.cpp
+++ b/src/gpu/cl/kernels/ClPool2dKernel.cpp
@@ -1,5 +1,5 @@
/*
- * Copyright (c) 2017-2021 Arm Limited.
+ * Copyright (c) 2017-2021, 2023 Arm Limited.
*
* SPDX-License-Identifier: MIT
*
@@ -182,7 +182,8 @@
}
else
{
- build_opts.add_option("-DINITIAL_VALUE=" + float_to_string_with_full_precision(std::numeric_limits<float>::lowest()));
+ std::string initial_value = pool_info.use_inf_as_limit ? "(-INFINITY)" : float_to_string_with_full_precision(std::numeric_limits<float>::lowest());
+ build_opts.add_option("-DINITIAL_VALUE=" + initial_value);
}
}
else