diff options
| author | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2022-01-20 16:33:29 +0000 |
|---|---|---|
| committer | Gian Marco Iodice <gianmarco.iodice@arm.com> | 2022-01-21 11:30:52 +0000 |
| commit | 0bae3eefbd77587c4495479ee4327433c06bf875 (patch) | |
| tree | 46619a1cde5c160ba303545c28de5c08a134cf33 | |
| parent | 066607f1384f502612869196c97b17ed0fc4caf3 (diff) | |
| download | ComputeLibrary-0bae3eefbd77587c4495479ee4327433c06bf875.tar.gz | |
Fix heuristic in ClConv2D
Resolves COMPMID-5040
Change-Id: I12f4c56da71ccee502f5656a80059f903877e8b9
Signed-off-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/6983
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
| -rw-r--r-- | src/gpu/cl/operators/ClConv2d.cpp | 20 |
1 files changed, 13 insertions, 7 deletions
diff --git a/src/gpu/cl/operators/ClConv2d.cpp b/src/gpu/cl/operators/ClConv2d.cpp index 2b9a51df22..3bb92de843 100644 --- a/src/gpu/cl/operators/ClConv2d.cpp +++ b/src/gpu/cl/operators/ClConv2d.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2021 Arm Limited. + * Copyright (c) 2021-2022 Arm Limited. * * SPDX-License-Identifier: MIT * @@ -255,9 +255,12 @@ ConvolutionMethod ClConv2d::get_convolution_method(const ITensorInfo *src, const // Floating-point case: GeMM/Direct/Winograd if(is_data_type_float(src->data_type())) { - const bool is_large_kernel_sz = (weights->dimension(idx_w) >= kernel_sz_direct_conv_thr) && (weights->dimension(idx_h) >= kernel_sz_direct_conv_thr); - const bool is_ifm_ge_16 = src->dimension(idx_c) >= 16; - const bool is_ifm_gt_ofm = weights->dimension(0U) * weights->dimension(1U) * weights->dimension(2U) > weights->dimension(3U); + // Get dst shape + TensorShape output_shape = misc::shape_calculator::compute_deep_convolution_shape(*src, *weights, conv_info); + const bool is_large_kernel_sz = (weights->dimension(idx_w) >= kernel_sz_direct_conv_thr) && (weights->dimension(idx_h) >= kernel_sz_direct_conv_thr); + const bool is_ifm_ge_16 = src->dimension(idx_c) >= 16; + const bool is_ofm_lte_8 = weights->dimension(3U) <= 8; + const bool workload_gte_8192 = (output_shape[0] * output_shape[1] * output_shape[2]) / 16 >= 8192; // Run Winograd if valid and IFM >= 16 if(is_wino_valid && is_ifm_ge_16) @@ -265,10 +268,13 @@ ConvolutionMethod ClConv2d::get_convolution_method(const ITensorInfo *src, const return ConvolutionMethod::WINOGRAD; } - // Run Direct for Large kernel size - if(is_large_kernel_sz && is_ifm_ge_16 && is_direct_valid && is_ifm_gt_ofm) + // Direct convolution case + if(is_direct_valid) { - return ConvolutionMethod::DIRECT; + if((is_large_kernel_sz && workload_gte_8192) || (is_ofm_lte_8 && is_ifm_ge_16)) + { + return ConvolutionMethod::DIRECT; + } } // Default case |