diff options
| author | Gunes Bayir <gunes.bayir@arm.com> | 2022-12-12 17:47:49 +0000 |
|---|---|---|
| committer | Gunes Bayir <gunes.bayir@arm.com> | 2022-12-14 09:15:34 +0000 |
| commit | a0ae8d2e6c57fd95c0edaf659b9df8b8c540d051 (patch) | |
| tree | 066dc0f7de97adc54a892ddfa4ed5e278f31f1be /src/runtime/CL | |
| parent | 32741725ac6e6c9658c51ed8585d314a1707ee8c (diff) | |
| download | ComputeLibrary-a0ae8d2e6c57fd95c0edaf659b9df8b8c540d051.tar.gz | |
Optimize Transposed Convolution for CL backend (Quantized)
This patch optimizes transposed convolution for QASYMM and QASYMM8_SIGNED types, by extending the transposed convolution kernel written for FP32/16.
Resolves: COMPMID-5723
Change-Id: Iab8f09231938adb949c506fd915ed45b885e5c7c
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8792
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Diffstat (limited to 'src/runtime/CL')
| -rw-r--r-- | src/runtime/CL/functions/CLDeconvolutionLayer.cpp | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/src/runtime/CL/functions/CLDeconvolutionLayer.cpp b/src/runtime/CL/functions/CLDeconvolutionLayer.cpp index a4db6d7770..ea7f3e75f7 100644 --- a/src/runtime/CL/functions/CLDeconvolutionLayer.cpp +++ b/src/runtime/CL/functions/CLDeconvolutionLayer.cpp @@ -141,14 +141,14 @@ DeconvolutionMethod CLDeconvolutionLayer::get_deconvolution_method(const ITensor { ARM_COMPUTE_UNUSED(output, bias, weights_info); - if(input->data_layout() == DataLayout::NHWC && (input->data_type() == DataType::F32 || input->data_type() == DataType::F16)) + if(is_data_type_quantized_per_channel(weights->data_type())) { - return DeconvolutionMethod::DIRECT; + return DeconvolutionMethod::UPSCALE_CONV2D; } - if(is_data_type_quantized_per_channel(weights->data_type())) + if(input->data_layout() == DataLayout::NHWC) { - return DeconvolutionMethod::UPSCALE_CONV2D; + return DeconvolutionMethod::DIRECT; } const DataLayout data_layout = input->data_layout(); |