/* * Copyright (c) 2016-2020 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 "arm_compute/core/NEON/kernels/NEScaleKernel.h" #include "arm_compute/core/AccessWindowStatic.h" #include "arm_compute/core/CPP/Validate.h" #include "arm_compute/core/Helpers.h" #include "arm_compute/core/NEON/wrapper/wrapper.h" #include "arm_compute/core/Window.h" #include "arm_compute/core/utils/misc/Rounding.h" #include "arm_compute/core/utils/misc/Utility.h" #include "src/core/utils/ScaleUtils.h" #include namespace arm_compute { namespace { Status validate_arguments(const ITensorInfo *input, const ITensorInfo *dx, const ITensorInfo *dy, const ITensorInfo *offsets, ITensorInfo *output, const ScaleKernelInfo &info) { ARM_COMPUTE_RETURN_ERROR_ON_CPU_F16_UNSUPPORTED(input); ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::U8, DataType::S16, DataType::F16, DataType::F32, DataType::QASYMM8, DataType::QASYMM8_SIGNED); ARM_COMPUTE_RETURN_ERROR_ON_NULLPTR(output); ARM_COMPUTE_RETURN_ERROR_ON_MISMATCHING_DATA_TYPES(input, output); ARM_COMPUTE_RETURN_ERROR_ON(output == input); ARM_COMPUTE_RETURN_ERROR_ON(info.sampling_policy != SamplingPolicy::CENTER && info.sampling_policy != SamplingPolicy::TOP_LEFT); ARM_COMPUTE_RETURN_ERROR_ON(!info.use_padding && info.border_mode != BorderMode::CONSTANT); ARM_COMPUTE_UNUSED(info.constant_border_value); const DataLayout data_layout = input->data_layout(); const auto width_index = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const auto height_index = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const auto output_width = output->dimension(width_index); const auto output_height = output->dimension(height_index); ARM_COMPUTE_RETURN_ERROR_ON(output_width == 0); ARM_COMPUTE_RETURN_ERROR_ON(output_height == 0); if(info.interpolation_policy == InterpolationPolicy::NEAREST_NEIGHBOR) { ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(offsets, 1, DataType::S32); } if(info.interpolation_policy == InterpolationPolicy::BILINEAR) { ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(offsets, 1, DataType::S32); ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(dx, 1, DataType::F32); ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(dy, 1, DataType::F32); } ARM_COMPUTE_RETURN_ERROR_ON(info.align_corners && !arm_compute::scale_utils::is_align_corners_allowed_sampling_policy(info.sampling_policy)); if(info.interpolation_policy == InterpolationPolicy::AREA) { ARM_COMPUTE_RETURN_ERROR_ON(data_layout != DataLayout::NCHW); ARM_COMPUTE_RETURN_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(input, 1, DataType::U8); } return Status{}; } std::pair validate_and_configure_window_nchw(ITensorInfo *input, ITensorInfo *dx, ITensorInfo *dy, ITensorInfo *offsets, ITensorInfo *output, const ScaleKernelInfo &info, BorderSize border_size) { bool window_changed{ false }; Window win{}; constexpr unsigned int num_elems_processed_per_iteration = 16; // Configure kernel window win = calculate_max_window(*output, Steps(num_elems_processed_per_iteration)); const ValidRegion &input_valid_region = input->valid_region(); if(offsets != nullptr) { AccessWindowHorizontal offsets_access(offsets, 0, num_elems_processed_per_iteration); window_changed = window_changed || update_window_and_padding(win, offsets_access); } if(dx != nullptr && dy != nullptr) { AccessWindowHorizontal dx_access(dx, 0, num_elems_processed_per_iteration); AccessWindowHorizontal dy_access(dy, 0, num_elems_processed_per_iteration); window_changed = window_changed || update_window_and_padding(win, dx_access, dy_access); } // Reads can occur within the valid region of the input AccessWindowStatic input_access(input, input_valid_region.anchor[0] - border_size.left, input_valid_region.anchor[1] - border_size.top, input_valid_region.anchor[0] + input_valid_region.shape[0] + border_size.right, input_valid_region.anchor[1] + input_valid_region.shape[1] + border_size.bottom); AccessWindowHorizontal output_access(output, 0, num_elems_processed_per_iteration); window_changed = window_changed || update_window_and_padding(win, input_access, output_access); output_access.set_valid_region(win, calculate_valid_region_scale(*input, output->tensor_shape(), info.interpolation_policy, info.sampling_policy, info.border_mode == BorderMode::UNDEFINED)); Status err = (window_changed) ? ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Insufficient Padding!") : Status{}; return std::make_pair(err, win); } std::pair validate_and_configure_window_nhwc(ITensorInfo *input, ITensorInfo *output, const ScaleKernelInfo &info, BorderSize border_size) { bool window_changed{ false }; Window win{}; const unsigned int num_elems_processed_per_iteration = (info.use_padding && info.interpolation_policy == InterpolationPolicy::NEAREST_NEIGHBOR) ? 16 / input->element_size() : 1; // Configure kernel window win = calculate_max_window(*output, Steps(num_elems_processed_per_iteration)); if(info.use_padding) { AccessWindowStatic input_access(input, 0, -border_size.top, ceil_to_multiple(input->tensor_shape()[0], num_elems_processed_per_iteration), input->tensor_shape()[1]); AccessWindowHorizontal output_access(output, 0, num_elems_processed_per_iteration); window_changed = update_window_and_padding(win, input_access, output_access); output->set_valid_region(calculate_valid_region_scale(*input, output->tensor_shape(), info.interpolation_policy, info.sampling_policy, info.border_mode == BorderMode::UNDEFINED)); } Status err = (window_changed) ? ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Insufficient Padding!") : Status{}; return std::make_pair(err, win); } std::pair validate_and_configure_window(ITensorInfo *input, ITensorInfo *dx, ITensorInfo *dy, ITensorInfo *offsets, ITensorInfo *output, const ScaleKernelInfo &info, BorderSize border_size) { std::pair win_config; switch(input->data_layout()) { case DataLayout::NCHW: if(!info.use_padding) { return std::make_pair(ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Padding required for NCHW"), Window{}); } win_config = validate_and_configure_window_nchw(input, dx, dy, offsets, output, info, border_size); break; case DataLayout::NHWC: win_config = validate_and_configure_window_nhwc(input, output, info, border_size); break; default: win_config = std::make_pair(ARM_COMPUTE_CREATE_ERROR(ErrorCode::RUNTIME_ERROR, "Unsupported data layout!"), Window{}); } return win_config; } template inline void scale_nearest_nhwc_core(const ITensor *input, const ITensor *offsets, ITensor *output, float hr, Window window, const Window &win_in, size_t stride_w, size_t stride_h, size_t stride_c, float sampling_offset, bool align_corners) { const int window_step_x = 16 / sizeof(T); const auto window_start_x = static_cast(window.x().start()); const auto window_end_x = static_cast(window.x().end()); window.set(Window::DimX, Window::Dimension(0, 1, 1)); Iterator in(input, win_in); Iterator out(output, window); const size_t offsets_stride = stride_w / sizeof(T); execute_window_loop(window, [&](const Coordinates & id) { const int32_t offset = *reinterpret_cast(offsets->ptr_to_element(Coordinates(id.y(), id.z()))); const auto in_yi = static_cast(align_corners ? arm_compute::utils::rounding::round_half_away_from_zero((id.z() + sampling_offset) * hr) : std::floor((id.z() + sampling_offset) * hr)); const int offset_row = in_yi * stride_h; int32_t x = window_start_x; for(; x < window_end_x - window_step_x; x += window_step_x) { wrapper::vstore(reinterpret_cast(out.ptr()) + x, wrapper::vloadq(reinterpret_cast(in.ptr() + offset * offsets_stride + offset_row + x * stride_c))); } for(; x < window_end_x; ++x) { *(reinterpret_cast(out.ptr()) + x) = *(reinterpret_cast(in.ptr() + offset * offsets_stride + offset_row + x * stride_c)); } }, in, out); } template inline void scale_bilinear_nhwc_core(const ITensor *input, const ITensor *offsets, const ITensor *dx, const ITensor *dy, ITensor *output, float hr, float sampling_offset, Window window, const Window &win_in, size_t stride_w, size_t stride_h, size_t stride_c, BorderMode border_mode, PixelValue constant_border_value, bool use_padding) { Iterator in(input, win_in); Iterator out(output, window); const size_t stride_w_elems = stride_w / sizeof(T); const size_t stride_h_elems = stride_h / sizeof(T); const int input_width = input->info()->dimension(1); const int input_height = input->info()->dimension(2); T border_value; if(use_padding && border_mode != BorderMode::REPLICATE) { // configure() sets top border to 0 for BorderMode::REPLICATE and border_value is not needed in execute_window_loop() for REPLICATE border_value = *reinterpret_cast(input->buffer() + input->info()->offset_first_element_in_bytes() - stride_w); } else { border_value = static_cast(constant_border_value.get()); } auto is_valid = [](int64_t x, int64_t low_x, int64_t high_x, int64_t y, int64_t low_y, int64_t high_y) { return !(x < low_x || x > high_x || y < low_y || y > high_y); }; int border_size = (border_mode == BorderMode::UNDEFINED) ? 0 : 1; const UniformQuantizationInfo iq_info = input->info()->quantization_info().uniform(); const UniformQuantizationInfo oq_info = output->info()->quantization_info().uniform(); execute_window_loop(window, [&](const Coordinates & id) { const auto offset = (*reinterpret_cast(offsets->ptr_to_element(Coordinates(id.y(), id.z())))) / static_cast(sizeof(T)); const auto dx_scale = *reinterpret_cast(dx->ptr_to_element(Coordinates(id.y(), id.z()))); const auto dy_scale = *reinterpret_cast(dy->ptr_to_element(Coordinates(id.y(), id.z()))); const int in_yi = std::floor((id.z() + sampling_offset) * hr - sampling_offset); const int offset_row = in_yi * stride_h + id.x() * stride_c; const T *in_ptr = reinterpret_cast(in.ptr() + offset * stride_w + offset_row); if(is_valid(offset, -border_size, input_width - 1 + border_size, in_yi, -border_size, input_height - 1 + border_size)) { T a00 = 0; T a01 = 0; T a10 = 0; T a11 = 0; if(border_mode == BorderMode::CONSTANT) { a00 = is_valid(offset, 0, input_width - 1, in_yi, 0, input_height - 1) ? *in_ptr : border_value; a01 = is_valid(offset + 1, 0, input_width - 1, in_yi, 0, input_height - 1) ? *(in_ptr + stride_w_elems) : border_value; a10 = is_valid(offset, 0, input_width - 1, in_yi + 1, 0, input_height - 1) ? *(in_ptr + stride_h_elems) : border_value; a11 = is_valid(offset + 1, 0, input_width - 1, in_yi + 1, 0, input_height - 1) ? *(in_ptr + stride_h_elems + stride_w_elems) : border_value; } else if(border_mode == BorderMode::REPLICATE) { auto clamped_x = utility::clamp(offset, 0, input_width - 1); auto clamped_x1 = utility::clamp(offset + 1, 0, input_width - 1); auto clamped_y = utility::clamp(in_yi, 0, input_height - 1); auto clamped_y1 = utility::clamp(in_yi + 1, 0, input_height - 1); a00 = *reinterpret_cast(in.ptr() + clamped_x * stride_w + clamped_y * stride_h + id.x() * stride_c); a01 = *reinterpret_cast(in.ptr() + clamped_x1 * stride_w + clamped_y * stride_h + id.x() * stride_c); a10 = *reinterpret_cast(in.ptr() + clamped_x * stride_w + clamped_y1 * stride_h + id.x() * stride_c); a11 = *reinterpret_cast(in.ptr() + clamped_x1 * stride_w + clamped_y1 * stride_h + id.x() * stride_c); } else { a00 = is_valid(offset, 0, input_width - 1, in_yi, 0, input_height - 1) ? *in_ptr : 0; a01 = is_valid(offset + 1, 0, input_width - 1, in_yi, 0, input_height - 1) ? *(in_ptr + stride_w_elems) : 0; a10 = is_valid(offset, 0, input_width - 1, in_yi + 1, 0, input_height - 1) ? *(in_ptr + stride_h_elems) : 0; a11 = is_valid(offset + 1, 0, input_width - 1, in_yi + 1, 0, input_height - 1) ? *(in_ptr + stride_h_elems + stride_w_elems) : 0; } // Perform interpolation const float dx1 = 1.0f - dx_scale; const float dy1 = 1.0f - dy_scale; const float w1 = dx1 * dy1; const float w2 = dx_scale * dy1; const float w3 = dx1 * dy_scale; const float w4 = dx_scale * dy_scale; T res = 0; //dequantize quantized input if(input->info()->data_type() == DataType::QASYMM8) { float inp00 = dequantize_qasymm8(a00, iq_info); float inp01 = dequantize_qasymm8(a01, iq_info); float inp10 = dequantize_qasymm8(a10, iq_info); float inp11 = dequantize_qasymm8(a11, iq_info); res = static_cast(quantize_qasymm8((inp00 * w1 + inp01 * w2 + inp10 * w3 + inp11 * w4), oq_info)); } else if(input->info()->data_type() == DataType::QASYMM8_SIGNED) { float inp00 = dequantize_qasymm8_signed(a00, iq_info); float inp01 = dequantize_qasymm8_signed(a01, iq_info); float inp10 = dequantize_qasymm8_signed(a10, iq_info); float inp11 = dequantize_qasymm8_signed(a11, iq_info); res = static_cast(quantize_qasymm8_signed((inp00 * w1 + inp01 * w2 + inp10 * w3 + inp11 * w4), oq_info)); } else { res = static_cast(a00 * w1 + a01 * w2 + a10 * w3 + a11 * w4); } // Store result *reinterpret_cast(out.ptr()) = res; } else { if(border_mode == BorderMode::CONSTANT) { *reinterpret_cast(out.ptr()) = border_value; } else if(border_mode == BorderMode::REPLICATE) { auto clamped_x = utility::clamp(offset, 0, input_width - 1); auto clamped_y = utility::clamp(in_yi, 0, input_height - 1); *reinterpret_cast(out.ptr()) = *reinterpret_cast(in.ptr() + clamped_x * stride_w + clamped_y * stride_h + id.x() * stride_c); } } }, in, out); } } // namespace NEScaleKernel::NEScaleKernel() : _func(nullptr), _offsets(nullptr), _dx(nullptr), _dy(nullptr), _input(nullptr), _output(nullptr), _policy(), _border_size(1), _border_mode(), _constant_border_value(PixelValue()), _sampling_offset(0), _use_padding(true), _align_corners(false) { } BorderSize NEScaleKernel::border_size() const { return _border_size; } void NEScaleKernel::configure(const ITensor *input, const ITensor *dx, const ITensor *dy, const ITensor *offsets, ITensor *output, const ScaleKernelInfo &info) { ARM_COMPUTE_ERROR_ON_NULLPTR(input, output); // Perform validation step ARM_COMPUTE_ERROR_THROW_ON(validate_arguments(input->info(), dx != nullptr ? dx->info() : nullptr, dy != nullptr ? dy->info() : nullptr, offsets != nullptr ? offsets->info() : nullptr, output->info(), info)); // Get data layout and width/height indices const DataLayout data_layout = input->info()->data_layout(); const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const int idx_height = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); _input = input; _output = output; _offsets = offsets; _dx = dx; _dy = dy; _policy = info.interpolation_policy; _border_size = BorderSize(1); _border_mode = info.border_mode; _constant_border_value = info.constant_border_value; _use_padding = info.use_padding; _align_corners = info.align_corners; if(info.sampling_policy == SamplingPolicy::CENTER) { _sampling_offset = 0.5f; } // Compute the ratio between source width/height and destination width/height const auto wr = arm_compute::scale_utils::calculate_resize_ratio(input->info()->dimension(idx_width), output->info()->dimension(idx_width), _align_corners); const auto hr = arm_compute::scale_utils::calculate_resize_ratio(input->info()->dimension(idx_height), output->info()->dimension(idx_height), _align_corners); // Add constant border only on top in case of NHWC layout if(data_layout == DataLayout::NHWC) { _border_size = (info.border_mode != BorderMode::REPLICATE && info.interpolation_policy == InterpolationPolicy::BILINEAR && info.use_padding) ? BorderSize(1, 0, 0, 0) : BorderSize(0); } // Area interpolation behaves as Nearest Neighbour in case of up-sampling const auto policy_to_use = (info.interpolation_policy == InterpolationPolicy::AREA && wr <= 1.f && hr <= 1.f) ? InterpolationPolicy::NEAREST_NEIGHBOR : _policy; // Select interpolation function switch(policy_to_use) { case InterpolationPolicy::NEAREST_NEIGHBOR: { _func = (data_layout == DataLayout::NCHW) ? &NEScaleKernel::scale_nearest_nchw : &NEScaleKernel::scale_nhwc; break; } case InterpolationPolicy::BILINEAR: { _func = (data_layout == DataLayout::NCHW) ? &NEScaleKernel::scale_bilinear_nchw : &NEScaleKernel::scale_nhwc; break; } case InterpolationPolicy::AREA: { _func = &NEScaleKernel::scale_area_nchw; break; } default: ARM_COMPUTE_ERROR("Unsupported interpolation mode"); } // Configure window std::pair win_config = validate_and_configure_window(input->info(), dx != nullptr ? dx->info() : nullptr, dy != nullptr ? dy->info() : nullptr, offsets != nullptr ? offsets->info() : nullptr, output->info(), info, border_size()); ARM_COMPUTE_ERROR_THROW_ON(win_config.first); INEKernel::configure(win_config.second); } void NEScaleKernel::scale_nearest_nchw(const Window &window) { const size_t input_stride = _input->info()->strides_in_bytes()[1]; // Compute the ratio between source height and destination height const auto hr = arm_compute::scale_utils::calculate_resize_ratio(_input->info()->dimension(1), _output->info()->dimension(1), _align_corners); // Don't increment in X and Y direction for the input tensor // A pointer to the start of this plane is needed as base for the precomputed offsets Window win_in(window); win_in.set(Window::DimX, Window::Dimension(0, 0, 0)); win_in.set(Window::DimY, Window::Dimension(0, 0, 0)); // Set offsets window Window win_off; win_off.set(Window::DimX, window[Window::DimX]); win_off.set(Window::DimY, window[Window::DimY]); for(size_t d = Window::DimZ; d < _offsets->info()->num_dimensions(); ++d) { win_off.set(d, Window::Dimension(0, 0, 0)); } // Create iterators Iterator in(_input, win_in); Iterator out(_output, window); Iterator offsets(_offsets, win_off); switch(_input->info()->data_type()) { case DataType::QASYMM8_SIGNED: { int8x16_t tmp = vdupq_n_s8(0); execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const uint8_t *const in_ptr = in.ptr(); const auto in_yi = static_cast(_align_corners ? arm_compute::utils::rounding::round_half_away_from_zero((id.y() + _sampling_offset) * hr) : std::floor((id.y() + _sampling_offset) * hr)); const int in_yi_clamped = std::min(static_cast(_input->info()->dimension(1)), std::max(in_yi, -1)); ARM_COMPUTE_ERROR_ON(in_yi_clamped < -1 || in_yi_clamped > static_cast(_input->info()->dimension(1))); const int offset_row = in_yi_clamped * input_stride; tmp = vsetq_lane_s8(in_ptr[offsets_ptr[0] + offset_row], tmp, 0); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[1] + offset_row], tmp, 1); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[2] + offset_row], tmp, 2); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[3] + offset_row], tmp, 3); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[4] + offset_row], tmp, 4); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[5] + offset_row], tmp, 5); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[6] + offset_row], tmp, 6); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[7] + offset_row], tmp, 7); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[8] + offset_row], tmp, 8); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[9] + offset_row], tmp, 9); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[10] + offset_row], tmp, 10); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[11] + offset_row], tmp, 11); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[12] + offset_row], tmp, 12); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[13] + offset_row], tmp, 13); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[14] + offset_row], tmp, 14); tmp = vsetq_lane_s8(in_ptr[offsets_ptr[15] + offset_row], tmp, 15); vst1q_s8(reinterpret_cast(out.ptr()), tmp); }, in, offsets, out); break; } case DataType::QASYMM8: case DataType::U8: { uint8x16_t tmp = vdupq_n_u8(0); execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const uint8_t *const in_ptr = in.ptr(); const auto in_yi = static_cast(_align_corners ? arm_compute::utils::rounding::round_half_away_from_zero((id.y() + _sampling_offset) * hr) : std::floor((id.y() + _sampling_offset) * hr)); const int in_yi_clamped = std::min(static_cast(_input->info()->dimension(1)), std::max(in_yi, -1)); ARM_COMPUTE_ERROR_ON(in_yi_clamped < -1 || in_yi_clamped > static_cast(_input->info()->dimension(1))); const int offset_row = in_yi_clamped * input_stride; tmp = vsetq_lane_u8(in_ptr[offsets_ptr[0] + offset_row], tmp, 0); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[1] + offset_row], tmp, 1); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[2] + offset_row], tmp, 2); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[3] + offset_row], tmp, 3); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[4] + offset_row], tmp, 4); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[5] + offset_row], tmp, 5); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[6] + offset_row], tmp, 6); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[7] + offset_row], tmp, 7); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[8] + offset_row], tmp, 8); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[9] + offset_row], tmp, 9); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[10] + offset_row], tmp, 10); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[11] + offset_row], tmp, 11); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[12] + offset_row], tmp, 12); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[13] + offset_row], tmp, 13); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[14] + offset_row], tmp, 14); tmp = vsetq_lane_u8(in_ptr[offsets_ptr[15] + offset_row], tmp, 15); vst1q_u8(out.ptr(), tmp); }, in, offsets, out); break; } case DataType::S16: { int16x8x2_t tmp = { { vdupq_n_s16(0), vdupq_n_s16(0) } }; execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto in_yi = static_cast(_align_corners ? arm_compute::utils::rounding::round_half_away_from_zero((id.y() + _sampling_offset) * hr) : std::floor((id.y() + _sampling_offset) * hr)); const int offset_row = in_yi * input_stride; tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[0] + offset_row), tmp.val[0], 0); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[2] + offset_row), tmp.val[0], 1); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[4] + offset_row), tmp.val[0], 2); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[6] + offset_row), tmp.val[0], 3); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[8] + offset_row), tmp.val[0], 4); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[10] + offset_row), tmp.val[0], 5); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[12] + offset_row), tmp.val[0], 6); tmp.val[0] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[14] + offset_row), tmp.val[0], 7); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[1] + offset_row), tmp.val[1], 0); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[3] + offset_row), tmp.val[1], 1); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[5] + offset_row), tmp.val[1], 2); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[7] + offset_row), tmp.val[1], 3); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[9] + offset_row), tmp.val[1], 4); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[11] + offset_row), tmp.val[1], 5); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[13] + offset_row), tmp.val[1], 6); tmp.val[1] = vsetq_lane_s16(*reinterpret_cast(in.ptr() + offsets_ptr[15] + offset_row), tmp.val[1], 7); vst2q_s16(reinterpret_cast(out.ptr()), tmp); }, in, offsets, out); break; } #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC case DataType::F16: { float16x8x2_t tmp = { { vdupq_n_f16(0), vdupq_n_f16(0) } }; execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto in_yi = static_cast(_align_corners ? arm_compute::utils::rounding::round_half_away_from_zero((id.y() + _sampling_offset) * hr) : std::floor((id.y() + _sampling_offset) * hr)); const int offset_row = in_yi * input_stride; tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[0] + offset_row), tmp.val[0], 0); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[2] + offset_row), tmp.val[0], 1); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[4] + offset_row), tmp.val[0], 2); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[6] + offset_row), tmp.val[0], 3); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[8] + offset_row), tmp.val[0], 4); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[10] + offset_row), tmp.val[0], 5); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[12] + offset_row), tmp.val[0], 6); tmp.val[0] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[14] + offset_row), tmp.val[0], 7); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[1] + offset_row), tmp.val[1], 0); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[3] + offset_row), tmp.val[1], 1); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[5] + offset_row), tmp.val[1], 2); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[7] + offset_row), tmp.val[1], 3); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[9] + offset_row), tmp.val[1], 4); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[11] + offset_row), tmp.val[1], 5); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[13] + offset_row), tmp.val[1], 6); tmp.val[1] = vsetq_lane_f16(*reinterpret_cast(in.ptr() + offsets_ptr[15] + offset_row), tmp.val[1], 7); vst2q_f16(reinterpret_cast<__fp16 *>(out.ptr()), tmp); }, in, offsets, out); break; } #endif /* __ARM_FEATURE_FP16_VECTOR_ARITHMETIC */ case DataType::F32: { float32x4x4_t tmp = { { vdupq_n_f32(0), vdupq_n_f32(0), vdupq_n_f32(0), vdupq_n_f32(0) } }; execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto in_yi = static_cast(_align_corners ? arm_compute::utils::rounding::round_half_away_from_zero((id.y() + _sampling_offset) * hr) : std::floor((id.y() + _sampling_offset) * hr)); const int offset_row = in_yi * input_stride; tmp.val[0] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[0] + offset_row), tmp.val[0], 0); tmp.val[0] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[4] + offset_row), tmp.val[0], 1); tmp.val[0] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[8] + offset_row), tmp.val[0], 2); tmp.val[0] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[12] + offset_row), tmp.val[0], 3); tmp.val[1] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[1] + offset_row), tmp.val[1], 0); tmp.val[1] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[5] + offset_row), tmp.val[1], 1); tmp.val[1] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[9] + offset_row), tmp.val[1], 2); tmp.val[1] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[13] + offset_row), tmp.val[1], 3); tmp.val[2] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[2] + offset_row), tmp.val[2], 0); tmp.val[2] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[6] + offset_row), tmp.val[2], 1); tmp.val[2] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[10] + offset_row), tmp.val[2], 2); tmp.val[2] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[14] + offset_row), tmp.val[2], 3); tmp.val[3] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[3] + offset_row), tmp.val[3], 0); tmp.val[3] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[7] + offset_row), tmp.val[3], 1); tmp.val[3] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[11] + offset_row), tmp.val[3], 2); tmp.val[3] = vsetq_lane_f32(*reinterpret_cast(in.ptr() + offsets_ptr[15] + offset_row), tmp.val[3], 3); vst4q_f32(reinterpret_cast(out.ptr()), tmp); }, in, offsets, out); break; } default: ARM_COMPUTE_ERROR("Not supported"); } } void NEScaleKernel::scale_bilinear_nchw(const Window &window) { ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(_input, 1, DataType::U8, DataType::QASYMM8, DataType::QASYMM8_SIGNED, DataType::S16, DataType::F16, DataType::F32); // Compute the ratio between source height and destination height const auto hr = arm_compute::scale_utils::calculate_resize_ratio(_input->info()->dimension(1), _output->info()->dimension(1), _align_corners); // Don't increment in X and Y direction for the input tensor // A pointer to the start of this plane is needed as base for the precomputed offsets Window win_in(window); win_in.set(Window::DimX, Window::Dimension(0, 0, 0)); win_in.set(Window::DimY, Window::Dimension(0, 0, 0)); Window win_off; win_off.set(Window::DimX, window.x()); win_off.set(Window::DimY, window.y()); for(size_t d = Window::DimZ; d < _offsets->info()->num_dimensions(); ++d) { win_off.set(d, Window::Dimension(0, 0, 0)); } Iterator in(_input, win_in); Iterator out(_output, window); Iterator offsets(_offsets, win_off); Iterator dx(_dx, win_off); Iterator dy(_dy, win_off); /* Input image stride */ const size_t in_stide_in_bytes = _input->info()->strides_in_bytes()[1]; const size_t in_stride = in_stide_in_bytes / _input->info()->element_size(); const UniformQuantizationInfo iq_info = _input->info()->quantization_info().uniform(); const UniformQuantizationInfo oq_info = _output->info()->quantization_info().uniform(); switch(_input->info()->data_type()) { case DataType::QASYMM8_SIGNED: { execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto dx_ptr = reinterpret_cast(dx.ptr()); const auto dy_ptr = reinterpret_cast(dy.ptr()); const auto in_ptr = reinterpret_cast(in.ptr()); const int in_yi = std::floor((id.y() + _sampling_offset) * hr - _sampling_offset); const int offset_row = in_yi * in_stide_in_bytes; int8x8_t tmp0 = vdup_n_s8(0); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[0] + offset_row], in_stride, dx_ptr[0], dy_ptr[0], iq_info, oq_info), tmp0, 0); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[1] + offset_row], in_stride, dx_ptr[1], dy_ptr[1], iq_info, oq_info), tmp0, 1); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[2] + offset_row], in_stride, dx_ptr[2], dy_ptr[2], iq_info, oq_info), tmp0, 2); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[3] + offset_row], in_stride, dx_ptr[3], dy_ptr[3], iq_info, oq_info), tmp0, 3); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[4] + offset_row], in_stride, dx_ptr[4], dy_ptr[4], iq_info, oq_info), tmp0, 4); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[5] + offset_row], in_stride, dx_ptr[5], dy_ptr[5], iq_info, oq_info), tmp0, 5); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[6] + offset_row], in_stride, dx_ptr[6], dy_ptr[6], iq_info, oq_info), tmp0, 6); tmp0 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[7] + offset_row], in_stride, dx_ptr[7], dy_ptr[7], iq_info, oq_info), tmp0, 7); int8x8_t tmp1 = vdup_n_s8(0); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[8] + offset_row], in_stride, dx_ptr[8], dy_ptr[8], iq_info, oq_info), tmp1, 0); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[9] + offset_row], in_stride, dx_ptr[9], dy_ptr[9], iq_info, oq_info), tmp1, 1); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[10] + offset_row], in_stride, dx_ptr[10], dy_ptr[10], iq_info, oq_info), tmp1, 2); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[11] + offset_row], in_stride, dx_ptr[11], dy_ptr[11], iq_info, oq_info), tmp1, 3); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[12] + offset_row], in_stride, dx_ptr[12], dy_ptr[12], iq_info, oq_info), tmp1, 4); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[13] + offset_row], in_stride, dx_ptr[13], dy_ptr[13], iq_info, oq_info), tmp1, 5); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[14] + offset_row], in_stride, dx_ptr[14], dy_ptr[14], iq_info, oq_info), tmp1, 6); tmp1 = vset_lane_s8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[15] + offset_row], in_stride, dx_ptr[15], dy_ptr[15], iq_info, oq_info), tmp1, 7); vst1q_s8(reinterpret_cast(out.ptr()), vcombine_s8(tmp0, tmp1)); }, in, offsets, dx, dy, out); break; } case DataType::QASYMM8: { execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto dx_ptr = reinterpret_cast(dx.ptr()); const auto dy_ptr = reinterpret_cast(dy.ptr()); const auto in_ptr = reinterpret_cast(in.ptr()); const int in_yi = std::floor((id.y() + _sampling_offset) * hr - _sampling_offset); const int offset_row = in_yi * in_stide_in_bytes; uint8x8_t tmp0 = vdup_n_u8(0); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[0] + offset_row], in_stride, dx_ptr[0], dy_ptr[0], iq_info, oq_info), tmp0, 0); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[1] + offset_row], in_stride, dx_ptr[1], dy_ptr[1], iq_info, oq_info), tmp0, 1); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[2] + offset_row], in_stride, dx_ptr[2], dy_ptr[2], iq_info, oq_info), tmp0, 2); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[3] + offset_row], in_stride, dx_ptr[3], dy_ptr[3], iq_info, oq_info), tmp0, 3); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[4] + offset_row], in_stride, dx_ptr[4], dy_ptr[4], iq_info, oq_info), tmp0, 4); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[5] + offset_row], in_stride, dx_ptr[5], dy_ptr[5], iq_info, oq_info), tmp0, 5); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[6] + offset_row], in_stride, dx_ptr[6], dy_ptr[6], iq_info, oq_info), tmp0, 6); tmp0 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[7] + offset_row], in_stride, dx_ptr[7], dy_ptr[7], iq_info, oq_info), tmp0, 7); uint8x8_t tmp1 = vdup_n_u8(0); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[8] + offset_row], in_stride, dx_ptr[8], dy_ptr[8], iq_info, oq_info), tmp1, 0); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[9] + offset_row], in_stride, dx_ptr[9], dy_ptr[9], iq_info, oq_info), tmp1, 1); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[10] + offset_row], in_stride, dx_ptr[10], dy_ptr[10], iq_info, oq_info), tmp1, 2); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[11] + offset_row], in_stride, dx_ptr[11], dy_ptr[11], iq_info, oq_info), tmp1, 3); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[12] + offset_row], in_stride, dx_ptr[12], dy_ptr[12], iq_info, oq_info), tmp1, 4); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[13] + offset_row], in_stride, dx_ptr[13], dy_ptr[13], iq_info, oq_info), tmp1, 5); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[14] + offset_row], in_stride, dx_ptr[14], dy_ptr[14], iq_info, oq_info), tmp1, 6); tmp1 = vset_lane_u8(delta_bilinear_c1_quantized(&in_ptr[offsets_ptr[15] + offset_row], in_stride, dx_ptr[15], dy_ptr[15], iq_info, oq_info), tmp1, 7); vst1q_u8(out.ptr(), vcombine_u8(tmp0, tmp1)); }, in, offsets, dx, dy, out); break; } case DataType::U8: { execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto dx_ptr = reinterpret_cast(dx.ptr()); const auto dy_ptr = reinterpret_cast(dy.ptr()); const auto in_ptr = reinterpret_cast(in.ptr()); const int in_yi = std::floor((id.y() + _sampling_offset) * hr - _sampling_offset); const int offset_row = in_yi * in_stide_in_bytes; uint8x8_t tmp0 = vdup_n_u8(0); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[0] + offset_row], in_stride, dx_ptr[0], dy_ptr[0]), tmp0, 0); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[1] + offset_row], in_stride, dx_ptr[1], dy_ptr[1]), tmp0, 1); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[2] + offset_row], in_stride, dx_ptr[2], dy_ptr[2]), tmp0, 2); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[3] + offset_row], in_stride, dx_ptr[3], dy_ptr[3]), tmp0, 3); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[4] + offset_row], in_stride, dx_ptr[4], dy_ptr[4]), tmp0, 4); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[5] + offset_row], in_stride, dx_ptr[5], dy_ptr[5]), tmp0, 5); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[6] + offset_row], in_stride, dx_ptr[6], dy_ptr[6]), tmp0, 6); tmp0 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[7] + offset_row], in_stride, dx_ptr[7], dy_ptr[7]), tmp0, 7); uint8x8_t tmp1 = vdup_n_u8(0); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[8] + offset_row], in_stride, dx_ptr[8], dy_ptr[8]), tmp1, 0); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[9] + offset_row], in_stride, dx_ptr[9], dy_ptr[9]), tmp1, 1); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[10] + offset_row], in_stride, dx_ptr[10], dy_ptr[10]), tmp1, 2); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[11] + offset_row], in_stride, dx_ptr[11], dy_ptr[11]), tmp1, 3); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[12] + offset_row], in_stride, dx_ptr[12], dy_ptr[12]), tmp1, 4); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[13] + offset_row], in_stride, dx_ptr[13], dy_ptr[13]), tmp1, 5); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[14] + offset_row], in_stride, dx_ptr[14], dy_ptr[14]), tmp1, 6); tmp1 = vset_lane_u8(delta_bilinear_c1(&in_ptr[offsets_ptr[15] + offset_row], in_stride, dx_ptr[15], dy_ptr[15]), tmp1, 7); vst1q_u8(out.ptr(), vcombine_u8(tmp0, tmp1)); }, in, offsets, dx, dy, out); break; } case DataType::S16: { execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto dx_ptr = reinterpret_cast(dx.ptr()); const auto dy_ptr = reinterpret_cast(dy.ptr()); const int in_yi = std::floor((id.y() + _sampling_offset) * hr - _sampling_offset); const int offset_row = in_yi * in_stide_in_bytes; int16x8x2_t tmp = { { vdupq_n_s16(0), vdupq_n_s16(0) } }; tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[0] + offset_row), in_stride, dx_ptr[0], dy_ptr[0]), tmp.val[0], 0); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[2] + offset_row), in_stride, dx_ptr[2], dy_ptr[2]), tmp.val[0], 1); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[4] + offset_row), in_stride, dx_ptr[4], dy_ptr[4]), tmp.val[0], 2); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[6] + offset_row), in_stride, dx_ptr[6], dy_ptr[6]), tmp.val[0], 3); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[8] + offset_row), in_stride, dx_ptr[8], dy_ptr[8]), tmp.val[0], 4); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[10] + offset_row), in_stride, dx_ptr[10], dy_ptr[10]), tmp.val[0], 5); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[12] + offset_row), in_stride, dx_ptr[12], dy_ptr[12]), tmp.val[0], 6); tmp.val[0] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[14] + offset_row), in_stride, dx_ptr[14], dy_ptr[14]), tmp.val[0], 7); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[1] + offset_row), in_stride, dx_ptr[1], dy_ptr[1]), tmp.val[1], 0); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[3] + offset_row), in_stride, dx_ptr[3], dy_ptr[3]), tmp.val[1], 1); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[5] + offset_row), in_stride, dx_ptr[5], dy_ptr[5]), tmp.val[1], 2); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[7] + offset_row), in_stride, dx_ptr[7], dy_ptr[7]), tmp.val[1], 3); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[9] + offset_row), in_stride, dx_ptr[9], dy_ptr[9]), tmp.val[1], 4); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[11] + offset_row), in_stride, dx_ptr[11], dy_ptr[11]), tmp.val[1], 5); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[13] + offset_row), in_stride, dx_ptr[13], dy_ptr[13]), tmp.val[1], 6); tmp.val[1] = vsetq_lane_s16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[15] + offset_row), in_stride, dx_ptr[15], dy_ptr[15]), tmp.val[1], 7); vst2q_s16(reinterpret_cast(out.ptr()), tmp); }, in, offsets, dx, dy, out); break; } #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC case DataType::F16: { execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto dx_ptr = reinterpret_cast(dx.ptr()); const auto dy_ptr = reinterpret_cast(dy.ptr()); const int in_yi = std::floor((id.y() + _sampling_offset) * hr - _sampling_offset); const int offset_row = in_yi * in_stide_in_bytes; float16x8x2_t tmp = { { vdupq_n_f16(0), vdupq_n_f16(0) } }; tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[0] + offset_row), in_stride, dx_ptr[0], dy_ptr[0]), tmp.val[0], 0); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[2] + offset_row), in_stride, dx_ptr[2], dy_ptr[2]), tmp.val[0], 1); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[4] + offset_row), in_stride, dx_ptr[4], dy_ptr[4]), tmp.val[0], 2); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[6] + offset_row), in_stride, dx_ptr[6], dy_ptr[6]), tmp.val[0], 3); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[8] + offset_row), in_stride, dx_ptr[8], dy_ptr[8]), tmp.val[0], 4); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[10] + offset_row), in_stride, dx_ptr[10], dy_ptr[10]), tmp.val[0], 5); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[12] + offset_row), in_stride, dx_ptr[12], dy_ptr[12]), tmp.val[0], 6); tmp.val[0] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[14] + offset_row), in_stride, dx_ptr[14], dy_ptr[14]), tmp.val[0], 7); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[1] + offset_row), in_stride, dx_ptr[1], dy_ptr[1]), tmp.val[1], 0); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[3] + offset_row), in_stride, dx_ptr[3], dy_ptr[3]), tmp.val[1], 1); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[5] + offset_row), in_stride, dx_ptr[5], dy_ptr[5]), tmp.val[1], 2); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[7] + offset_row), in_stride, dx_ptr[7], dy_ptr[7]), tmp.val[1], 3); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[9] + offset_row), in_stride, dx_ptr[9], dy_ptr[9]), tmp.val[1], 4); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[11] + offset_row), in_stride, dx_ptr[11], dy_ptr[11]), tmp.val[1], 5); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[13] + offset_row), in_stride, dx_ptr[13], dy_ptr[13]), tmp.val[1], 6); tmp.val[1] = vsetq_lane_f16(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[15] + offset_row), in_stride, dx_ptr[15], dy_ptr[15]), tmp.val[1], 7); vst2q_f16(reinterpret_cast<__fp16 *>(out.ptr()), tmp); }, in, offsets, dx, dy, out); break; } #endif /* __ARM_FEATURE_FP16_VECTOR_ARITHMETIC */ case DataType::F32: { execute_window_loop(window, [&](const Coordinates & id) { const auto offsets_ptr = reinterpret_cast(offsets.ptr()); const auto dx_ptr = reinterpret_cast(dx.ptr()); const auto dy_ptr = reinterpret_cast(dy.ptr()); const int in_yi = std::floor((id.y() + _sampling_offset) * hr - _sampling_offset); const int offset_row = in_yi * in_stide_in_bytes; float32x4x4_t tmp = { { vdupq_n_f32(0), vdupq_n_f32(0), vdupq_n_f32(0), vdupq_n_f32(0) } }; tmp.val[0] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[0] + offset_row), in_stride, dx_ptr[0], dy_ptr[0]), tmp.val[0], 0); tmp.val[0] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[4] + offset_row), in_stride, dx_ptr[4], dy_ptr[4]), tmp.val[0], 1); tmp.val[0] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[8] + offset_row), in_stride, dx_ptr[8], dy_ptr[8]), tmp.val[0], 2); tmp.val[0] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[12] + offset_row), in_stride, dx_ptr[12], dy_ptr[12]), tmp.val[0], 3); tmp.val[1] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[1] + offset_row), in_stride, dx_ptr[1], dy_ptr[1]), tmp.val[1], 0); tmp.val[1] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[5] + offset_row), in_stride, dx_ptr[5], dy_ptr[5]), tmp.val[1], 1); tmp.val[1] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[9] + offset_row), in_stride, dx_ptr[9], dy_ptr[9]), tmp.val[1], 2); tmp.val[1] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[13] + offset_row), in_stride, dx_ptr[13], dy_ptr[13]), tmp.val[1], 3); tmp.val[2] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[2] + offset_row), in_stride, dx_ptr[2], dy_ptr[2]), tmp.val[2], 0); tmp.val[2] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[6] + offset_row), in_stride, dx_ptr[6], dy_ptr[6]), tmp.val[2], 1); tmp.val[2] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[10] + offset_row), in_stride, dx_ptr[10], dy_ptr[10]), tmp.val[2], 2); tmp.val[2] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[14] + offset_row), in_stride, dx_ptr[14], dy_ptr[14]), tmp.val[2], 3); tmp.val[3] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[3] + offset_row), in_stride, dx_ptr[3], dy_ptr[3]), tmp.val[3], 0); tmp.val[3] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[7] + offset_row), in_stride, dx_ptr[7], dy_ptr[7]), tmp.val[3], 1); tmp.val[3] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[11] + offset_row), in_stride, dx_ptr[11], dy_ptr[11]), tmp.val[3], 2); tmp.val[3] = vsetq_lane_f32(delta_bilinear_c1(reinterpret_cast(in.ptr() + offsets_ptr[15] + offset_row), in_stride, dx_ptr[15], dy_ptr[15]), tmp.val[3], 3); vst4q_f32(reinterpret_cast(out.ptr()), tmp); }, in, offsets, dx, dy, out); break; } default: ARM_COMPUTE_ERROR("Not supported"); break; } } void NEScaleKernel::scale_area_nchw(const Window &window) { ARM_COMPUTE_ERROR_ON_DATA_TYPE_CHANNEL_NOT_IN(_input, 1, DataType::U8); // Don't increment in width/height/channels for the input tensor // A pointer to the start of this plane is needed as base for the precomputed offsets Window win_in(window); win_in.set(Window::DimX, Window::Dimension(0, 0, 0)); win_in.set(Window::DimY, Window::Dimension(0, 0, 0)); win_in.set(Window::DimZ, Window::Dimension(0, 0, 0)); Iterator in(_input, win_in); Iterator out(_output, window); const auto wr = arm_compute::scale_utils::calculate_resize_ratio(_input->info()->dimension(0), _output->info()->dimension(0), _align_corners); const auto hr = arm_compute::scale_utils::calculate_resize_ratio(_input->info()->dimension(1), _output->info()->dimension(1), _align_corners); const auto w = _input->info()->dimension(0); const auto h = _input->info()->dimension(1); const size_t in_stride = _input->info()->strides_in_bytes()[1]; execute_window_loop(window, [&](const Coordinates & id) { const auto in_ptr = reinterpret_cast(in.ptr()); uint8x8_t tmp0 = vdup_n_u8(0); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x(), id.y()), tmp0, 0); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 1, id.y()), tmp0, 1); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 2, id.y()), tmp0, 2); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 3, id.y()), tmp0, 3); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 4, id.y()), tmp0, 4); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 5, id.y()), tmp0, 5); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 6, id.y()), tmp0, 6); tmp0 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 7, id.y()), tmp0, 7); uint8x8_t tmp1 = vdup_n_u8(0); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 8, id.y()), tmp1, 0); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 9, id.y()), tmp1, 1); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 10, id.y()), tmp1, 2); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 11, id.y()), tmp1, 3); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 12, id.y()), tmp1, 4); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 13, id.y()), tmp1, 5); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 14, id.y()), tmp1, 6); tmp1 = vset_lane_u8(pixel_area_c1u8_clamp(in_ptr, in_stride, w, h, wr, hr, id.x() + 15, id.y()), tmp1, 7); vst1q_u8(out.ptr(), vcombine_u8(tmp0, tmp1)); }, in, out); } void NEScaleKernel::scale_nhwc(const Window &window) { // Get data layout and width/height indices const DataLayout data_layout = DataLayout::NHWC; const int idx_channels = get_data_layout_dimension_index(data_layout, DataLayoutDimension::CHANNEL); const int idx_width = get_data_layout_dimension_index(data_layout, DataLayoutDimension::WIDTH); const int idx_height = get_data_layout_dimension_index(data_layout, DataLayoutDimension::HEIGHT); const size_t input_stride_w = _input->info()->strides_in_bytes()[idx_width]; const size_t input_stride_h = _input->info()->strides_in_bytes()[idx_height]; const size_t input_stride_c = _input->info()->strides_in_bytes()[idx_channels]; // Compute the ratio between source height and destination height const auto hr = arm_compute::scale_utils::calculate_resize_ratio(_input->info()->dimension(idx_height), _output->info()->dimension(idx_height), _align_corners); // Don't increment in width/height/channels for the input tensor // A pointer to the start of this plane is needed as base for the precomputed offsets Window win_in(window); win_in.set(Window::DimX, Window::Dimension(0, 0, 0)); win_in.set(Window::DimY, Window::Dimension(0, 0, 0)); win_in.set(Window::DimZ, Window::Dimension(0, 0, 0)); switch(_input->info()->data_type()) { case DataType::QASYMM8_SIGNED: { if(_policy == InterpolationPolicy::NEAREST_NEIGHBOR) { scale_nearest_nhwc_core(_input, _offsets, _output, hr, window, win_in, input_stride_w, input_stride_h, input_stride_c, _sampling_offset, _align_corners); } else { scale_bilinear_nhwc_core(_input, _offsets, _dx, _dy, _output, hr, _sampling_offset, window, win_in, input_stride_w, input_stride_h, input_stride_c, _border_mode, _constant_border_value, _use_padding); } break; } case DataType::QASYMM8: case DataType::U8: { if(_policy == InterpolationPolicy::NEAREST_NEIGHBOR) { scale_nearest_nhwc_core(_input, _offsets, _output, hr, window, win_in, input_stride_w, input_stride_h, input_stride_c, _sampling_offset, _align_corners); } else { scale_bilinear_nhwc_core(_input, _offsets, _dx, _dy, _output, hr, _sampling_offset, window, win_in, input_stride_w, input_stride_h, input_stride_c, _border_mode, _constant_border_value, _use_padding); } break; } case DataType::S16: { if(_policy == InterpolationPolicy::NEAREST_NEIGHBOR) { scale_nearest_nhwc_core(_input, _offsets, _output, hr, window, win_in, input_stride_w, input_stride_h, input_stride_c, _sampling_offset, _align_corners); } else { scale_bilinear_nhwc_core(_input, _offsets, _dx, _dy, _output, hr, _sampling_offset, window, win_in, input_stride_w, input_stride_h, input_stride_c, _border_mode, _constant_border_value, _use_padding); } break; } #ifdef __ARM_FEATURE_FP16_VECTOR_ARITHMETIC case DataType::F16: { if(_policy == InterpolationPolicy::NEAREST_NEIGHBOR) { scale_nearest_nhwc_core(_input, _offsets, _output, hr, window, win_in, input_stride_w, input_stride_h, input_stride_c, _sampling_offset, _align_corners); } else { scale_bilinear_nhwc_core(_input, _offsets, _dx, _dy, _output, hr, _sampling_offset, window, win_in, input_stride_w, input_stride_h, input_stride_c, _border_mode, _constant_border_value, _use_padding); } break; } #endif /* __ARM_FEATURE_FP16_VECTOR_ARITHMETIC */ case DataType::F32: { if(_policy == InterpolationPolicy::NEAREST_NEIGHBOR) { scale_nearest_nhwc_core(_input, _offsets, _output, hr, window, win_in, input_stride_w, input_stride_h, input_stride_c, _sampling_offset, _align_corners); } else { scale_bilinear_nhwc_core(_input, _offsets, _dx, _dy, _output, hr, _sampling_offset, window, win_in, input_stride_w, input_stride_h, input_stride_c, _border_mode, _constant_border_value, _use_padding); } break; } default: ARM_COMPUTE_ERROR("Not supported"); break; } } Status NEScaleKernel::validate(const ITensorInfo *input, const ITensorInfo *dx, const ITensorInfo *dy, const ITensorInfo *offsets, ITensorInfo *output, const ScaleKernelInfo &info) { BorderSize border_size(1); if(input->data_layout() == DataLayout::NHWC) { border_size = (info.border_mode == BorderMode::CONSTANT && info.interpolation_policy == InterpolationPolicy::BILINEAR) ? BorderSize(1, 0, 0, 0) : BorderSize(0); } ARM_COMPUTE_RETURN_ON_ERROR(validate_arguments(input, dx, dy, offsets, output, info)); ARM_COMPUTE_RETURN_ON_ERROR(validate_and_configure_window(input->clone().get(), dx != nullptr ? dx->clone().get() : nullptr, dy != nullptr ? dy->clone().get() : nullptr, offsets != nullptr ? offsets->clone().get() : nullptr, output->clone().get(), info, border_size) .first); return Status{}; } void NEScaleKernel::run(const Window &window, const ThreadInfo &info) { ARM_COMPUTE_UNUSED(info); ARM_COMPUTE_ERROR_ON_UNCONFIGURED_KERNEL(this); ARM_COMPUTE_ERROR_ON_INVALID_SUBWINDOW(INEKernel::window(), window); ARM_COMPUTE_ERROR_ON(_func == nullptr); (this->*_func)(window); } } // namespace arm_compute