/* * Copyright (c) 2022 Arm Limited. * * SPDX-License-Identifier: MIT * * Permission is hereby granted, free of charge, to any person obtaining a copy * of this software and associated documentation files (the "Software"), to * deal in the Software without restriction, including without limitation the * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or * sell copies of the Software, and to permit persons to whom the Software is * furnished to do so, subject to the following conditions: * * The above copyright notice and this permission notice shall be included in all * copies or substantial portions of the Software. * * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE * SOFTWARE. */ #include "Pooling3dLayer.h" #include "arm_compute/core/utils/misc/ShapeCalculator.h" #include "tests/validation/Helpers.h" namespace arm_compute { namespace test { namespace validation { namespace reference { using namespace arm_compute::misc::shape_calculator; template SimpleTensor pooling_3d_layer_internal(const SimpleTensor &src, const Pooling3dLayerInfo &pool3d_info, SimpleTensor *indices) { TensorShape pooled_shape = compute_pool3d_shape(src.shape(), pool3d_info); SimpleTensor dst{ pooled_shape, src.data_type(), 1 }; if(indices != nullptr) { *indices = SimpleTensor { pooled_shape, DataType::U32, 1 }; } const int idx_channel = 0; const int idx_width = 1; const int idx_height = 2; const int idx_depth = 3; const int idx_batch = 4; const int pool_size_width = pool3d_info.is_global_pooling ? src.shape()[idx_width] : pool3d_info.pool_size.width; const int pool_size_height = pool3d_info.is_global_pooling ? src.shape()[idx_height] : pool3d_info.pool_size.height; const int pool_size_depth = pool3d_info.is_global_pooling ? src.shape()[idx_depth] : pool3d_info.pool_size.depth; const int pool_stride_width = static_cast(pool3d_info.stride.width); const int pool_stride_height = static_cast(pool3d_info.stride.height); const int pool_stride_depth = static_cast(pool3d_info.stride.depth); const int pad_left = static_cast(pool3d_info.padding.left); const int pad_top = static_cast(pool3d_info.padding.top); const int pad_front = static_cast(pool3d_info.padding.front); const int pad_right = static_cast(pool3d_info.padding.right); const int pad_bottom = static_cast(pool3d_info.padding.bottom); const int pad_back = static_cast(pool3d_info.padding.back); const int num_channels = static_cast(src.shape()[idx_channel]); const int num_batches = static_cast(src.shape()[idx_batch]); ARM_COMPUTE_ERROR_ON(num_channels != static_cast(dst.shape()[idx_channel])); ARM_COMPUTE_ERROR_ON(num_batches != static_cast(dst.shape()[idx_batch])); const int w_src = static_cast(src.shape()[idx_width]); const int h_src = static_cast(src.shape()[idx_height]); const int d_src = static_cast(src.shape()[idx_depth]); const int w_dst = static_cast(dst.shape()[idx_width]); const int h_dst = static_cast(dst.shape()[idx_height]); const int d_dst = static_cast(dst.shape()[idx_depth]); const bool exclude_padding = pool3d_info.exclude_padding; const int height_stride_src = num_channels * w_src; const int depth_stride_src = height_stride_src * h_src; const int batch_stride_src = depth_stride_src * d_src; const int height_stride_dst = num_channels * w_dst; const int depth_stride_dst = height_stride_dst * h_dst; const int batch_stride_dst = depth_stride_dst * d_dst; for(int b = 0; b < num_batches; ++b) { const int batch_offset_dst = b * batch_stride_dst; const int batch_offset_src = b * batch_stride_src; for(int c = 0; c < num_channels; ++c) { for(int d = 0; d < d_dst; ++d) { const int depth_offset_dst = d * depth_stride_dst; for(int h = 0; h < h_dst; ++h) { const int height_offset_dst = h * height_stride_dst; for(int w = 0; w < w_dst; ++w) { int wstart = w * pool_stride_width - pad_left; int hstart = h * pool_stride_height - pad_top; int dstart = d * pool_stride_depth - pad_front; int wend = std::min(wstart + pool_size_width, w_src + pad_right); int hend = std::min(hstart + pool_size_height, h_src + pad_bottom); int dend = std::min(dstart + pool_size_depth, d_src + pad_back); // this may not be equal to pool_w * pool_h * pool_d because of // DimensionRoundingType choice (CEIL) int pool_size = (dend - dstart) * (hend - hstart) * (wend - wstart); // limit [start, end) to [0, w_src) wstart = std::max(wstart, 0); hstart = std::max(hstart, 0); dstart = std::max(dstart, 0); wend = std::min(wend, w_src); hend = std::min(hend, h_src); dend = std::min(dend, d_src); auto max_val = -std::numeric_limits::infinity(); int max_index{ 0 }; T avg_val = static_cast(0.f); T l2_val = static_cast(0.f); if(exclude_padding) { pool_size = (dend - dstart) * (hend - hstart) * (wend - wstart); } for(int z = dstart; z < dend; ++z) { const int depth_offset_src = z * depth_stride_src; for(int y = hstart; y < hend; ++y) { const int height_offset_src = y * height_stride_src; for(int x = wstart; x < wend; ++x) { const auto val = static_cast( src[batch_offset_src + depth_offset_src + height_offset_src + x * num_channels + c]); if(val > max_val) { max_val = val; max_index = coord2index(src.shape(), Coordinates(c, x, y, z, 0)); } avg_val += val; l2_val += val * val; } } } avg_val /= pool_size; l2_val = static_cast(std::sqrt(l2_val / pool_size)); int dst_index = batch_offset_dst + depth_offset_dst + height_offset_dst + w * num_channels + c; switch(pool3d_info.pool_type) { case PoolingType::MAX: dst[dst_index] = static_cast(max_val); break; case PoolingType::AVG: dst[dst_index] = static_cast(avg_val); break; case PoolingType::L2: dst[dst_index] = static_cast(l2_val); break; default: ARM_COMPUTE_ERROR("Pooling Type should be either MAX, AVG or L2"); } if(indices != nullptr) { (*indices)[dst_index] = max_index; } } } } } } return dst; } template SimpleTensor pooling_3d_layer(const SimpleTensor &src, const Pooling3dLayerInfo &pool3d_info, const QuantizationInfo &output_qinfo, SimpleTensor *indices); template SimpleTensor pooling_3d_layer(const SimpleTensor &src, const Pooling3dLayerInfo &pool3d_info, const QuantizationInfo &output_qinfo, SimpleTensor *indices); template SimpleTensor pooling_3d_layer(const SimpleTensor &src, const Pooling3dLayerInfo &pool3d_info, const QuantizationInfo &output_qinfo, SimpleTensor *indices) { ARM_COMPUTE_UNUSED(output_qinfo); return pooling_3d_layer_internal(src, pool3d_info, indices); } template <> SimpleTensor pooling_3d_layer(const SimpleTensor &src, const Pooling3dLayerInfo &pool3d_info, const QuantizationInfo &output_qinfo, SimpleTensor *indices) { SimpleTensor src_tmp = convert_from_asymmetric(src); SimpleTensor dst_tmp = pooling_3d_layer_internal(src_tmp, pool3d_info, indices); return convert_to_asymmetric(dst_tmp, output_qinfo); } template <> SimpleTensor pooling_3d_layer(const SimpleTensor &src, const Pooling3dLayerInfo &pool3d_info, const QuantizationInfo &output_qinfo, SimpleTensor *indices) { SimpleTensor src_tmp = convert_from_asymmetric(src); SimpleTensor dst_tmp = pooling_3d_layer_internal(src_tmp, pool3d_info, indices); return convert_to_asymmetric(dst_tmp, output_qinfo); } } // namespace reference } // namespace validation } // namespace test } // namespace arm_compute