Add Resize/Scale operator to Dynamic Fusion interface
Resolves: COMPMID-5521
Change-Id: Id38a4ce18f9ea8805a151acb064e72795535d1a0
Signed-off-by: Jakub Sujak <jakub.sujak@arm.com>
Signed-off-by: Gunes Bayir <gunes.bayir@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/8859
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>
diff --git a/src/dynamic_fusion/sketch/gpu/template_writer/cl/ClTemplateResize.cpp b/src/dynamic_fusion/sketch/gpu/template_writer/cl/ClTemplateResize.cpp
new file mode 100644
index 0000000..7ee79e8
--- /dev/null
+++ b/src/dynamic_fusion/sketch/gpu/template_writer/cl/ClTemplateResize.cpp
@@ -0,0 +1,310 @@
+/*
+ * 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 "ClTemplateResize.h"
+
+#include "src/core/helpers/WindowHelpers.h"
+#include "src/core/utils/ScaleUtils.h"
+#include "src/dynamic_fusion/sketch/gpu/GpuKernelComponentGroup.h"
+
+namespace arm_compute
+{
+namespace experimental
+{
+namespace dynamic_fusion
+{
+ClTemplateResize::ClTemplateResize(ComponentId id, const ArgumentPack<ITensorInfo> &tensors, const ClTemplateResize::Attributes &attributes)
+ : IGpuTemplateComponentWriter{ id, tensors }, _src{}, _dst{}, _attributes{ attributes }
+{
+ _src = this->tensors().get_const_tensor(TensorType::ACL_SRC_0);
+ _dst = this->tensors().get_const_tensor(TensorType::ACL_DST_0);
+
+ ARM_COMPUTE_ERROR_ON_NULLPTR(_src, _dst);
+}
+
+std::string ClTemplateResize::get_name() const
+{
+ return _attributes.interpolation_policy() == InterpolationPolicy::BILINEAR ? "resize_bilinear" : "resize_nearest";
+}
+
+std::string ClTemplateResize::get_component_code(const IGpuTemplateComponentWriter::ComponentGroup &comp_group) const
+{
+ ARM_COMPUTE_UNUSED(comp_group);
+
+ std::string code = R"_(
+//------------------ START KERNEL {{meta_kernel_id}} ---------------------
+TILE({{DST_DATA_TYPE}}, 1, N0, {{dst}});
+TILE(uint, 1, 1, g_dst_indirect_y);
+{
+ const int yo = g_ind_2 % {{arg_dst}}_h;
+ const int bout = g_ind_2 / {{arg_dst}}_h;
+)_";
+
+ if(_attributes.interpolation_policy() == InterpolationPolicy::NEAREST_NEIGHBOR)
+ {
+ if(_attributes.sampling_policy() == SamplingPolicy::TOP_LEFT)
+ {
+ code += R"_(
+ float xi_f = (g_ind_1 * SCALE_X);
+ float yi_f = (yo * SCALE_Y);
+)_";
+ }
+ else
+ {
+ code += R"_(
+ float xi_f = ((g_ind_1 + 0.5f) * SCALE_X);
+ float yi_f = ((yo + 0.5f) * SCALE_Y);
+)_";
+ }
+
+ if(_attributes.align_corners())
+ {
+ code += R"_(
+ xi_f = round(xi_f);
+ yi_f = round(yi_f);
+)_";
+ }
+
+ code += R"_(
+ const int xi0 = clamp((int)xi_f, 0, (int){{src}}_w - 1);
+ const int yi0 = clamp((int)yi_f, 0, (int){{src}}_h - 1);
+
+ T_LOAD_NHWC_WITH_DILATION({{SRC_DATA_TYPE}}, 1, 1, N0, {{SRC_TENSOR_TYPE}}, {{src}}, bout, yi0, xi0, g_ind_0, {{src}}_w, {{src}}_h, 1, 1, false, {{dst}});
+)_";
+ }
+ else if(_attributes.interpolation_policy() == InterpolationPolicy::BILINEAR)
+ {
+ if(_attributes.sampling_policy() == SamplingPolicy::TOP_LEFT)
+ {
+ code += R"_(
+ float xi_f = (g_ind_1 * SCALE_X);
+ float yi_f = (yo * SCALE_Y);
+)_";
+ }
+ else
+ {
+ code += R"_(
+ float xi_f = ((g_ind_1 + 0.5f) * SCALE_X - 0.5f);
+ float yi_f = ((yo + 0.5f) * SCALE_Y - 0.5f);
+)_";
+ }
+
+ code += R"_(
+ const int xi = (int)floor(xi_f);
+ const int yi = (int)floor(yi_f);
+
+ TILE({{SRC_DATA_TYPE}}, 1, N0, in00);
+ TILE({{SRC_DATA_TYPE}}, 1, N0, in01);
+ TILE({{SRC_DATA_TYPE}}, 1, N0, in10);
+ TILE({{SRC_DATA_TYPE}}, 1, N0, in11);
+
+ in00[0].v = {{CONSTANT_VALUE}};
+ in01[0].v = {{CONSTANT_VALUE}};
+ in10[0].v = {{CONSTANT_VALUE}};
+ in11[0].v = {{CONSTANT_VALUE}};
+
+ const int xi0 = clamp(xi, 0, (int){{src}}_w - 1);
+ const int yi0 = clamp(yi, 0, (int){{src}}_h - 1);
+ const int xi1 = clamp(xi + 1, 0, (int){{src}}_w - 1);
+ const int yi1 = clamp(yi + 1, 0, (int){{src}}_h - 1);
+
+ T_LOAD_NHWC_WITH_DILATION({{SRC_DATA_TYPE}}, 1, 1, N0, {{SRC_TENSOR_TYPE}}, {{src}}, bout, yi0, xi0, g_ind_0, {{src}}_w, {{src}}_h, 1, 1, false, in00);
+ T_LOAD_NHWC_WITH_DILATION({{SRC_DATA_TYPE}}, 1, 1, N0, {{SRC_TENSOR_TYPE}}, {{src}}, bout, yi0, xi1, g_ind_0, {{src}}_w, {{src}}_h, 1, 1, false, in01);
+ T_LOAD_NHWC_WITH_DILATION({{SRC_DATA_TYPE}}, 1, 1, N0, {{SRC_TENSOR_TYPE}}, {{src}}, bout, yi1, xi0, g_ind_0, {{src}}_w, {{src}}_h, 1, 1, false, in10);
+ T_LOAD_NHWC_WITH_DILATION({{SRC_DATA_TYPE}}, 1, 1, N0, {{SRC_TENSOR_TYPE}}, {{src}}, bout, yi1, xi1, g_ind_0, {{src}}_w, {{src}}_h, 1, 1, false, in11);
+)_";
+
+ if(is_data_type_float(_src->data_type()))
+ {
+ code += R"_(
+ const {{SRC_DATA_TYPE}} a = ({{SRC_DATA_TYPE}})(xi_f - (float)xi);
+ const {{SRC_DATA_TYPE}} b = ({{SRC_DATA_TYPE}})(1.f - a);
+ const {{SRC_DATA_TYPE}} a1 = ({{SRC_DATA_TYPE}})(yi_f - (float)yi);
+ const {{SRC_DATA_TYPE}} b1 = ({{SRC_DATA_TYPE}})(1.f - a1);
+
+ // Calculate the output
+ {{dst}}[0].v = ((in00[0].v * b * b1) + (in01[0].v * a * b1) + (in10[0].v * b * a1) + (in11[0].v * a * a1));
+)_";
+ }
+ else
+ {
+ code += R"_(
+ TILE(float, 1, N0, out_f);
+ TILE(float, 1, N0, in00_f);
+ TILE(float, 1, N0, in01_f);
+ TILE(float, 1, N0, in10_f);
+ TILE(float, 1, N0, in11_f);
+
+ const float a = (xi_f - (float)xi);
+ const float b = (1.f - a);
+ const float a1 = (yi_f - (float)yi);
+ const float b1 = (1.f - a1);
+)_"
+ // Dequantize
+ R"_(
+ LOOP_UNROLLING(int, n0, 0, 1, N0,
+ {
+ in00_f[0].s[n0] = ((float)in00[0].s[n0] - (float){{OFFSET}}) * (float){{SCALE}};
+ in01_f[0].s[n0] = ((float)in01[0].s[n0] - (float){{OFFSET}}) * (float){{SCALE}};
+ in10_f[0].s[n0] = ((float)in10[0].s[n0] - (float){{OFFSET}}) * (float){{SCALE}};
+ in11_f[0].s[n0] = ((float)in11[0].s[n0] - (float){{OFFSET}}) * (float){{SCALE}};
+ })
+)_"
+ // Calculate the output in the floating-point domain
+ R"_(
+ out_f[0].v = ((in00_f[0].v * b * b1) + (in01_f[0].v * a * b1) + (in10_f[0].v * b * a1) + (in11_f[0].v * a * a1));
+)_"
+ // Quantize
+ R"_(
+ LOOP_UNROLLING(int, n0, 0, 1, N0,
+ {
+ {{dst}}[0].s[n0] = CONVERT_SAT(out_f[0].s[n0] / (float){{SCALE}} + (float){{OFFSET}}, {{DST_DATA_TYPE}});
+ })
+)_";
+ }
+ }
+ else
+ {
+ ARM_COMPUTE_ERROR("Unsupported interpolation policy");
+ }
+
+ code += R"_(
+ g_dst_indirect_y[0].v = g_ind_1 + (yo * (int)({{arg_dst}}_w)) + bout * (int)({{arg_dst}}_w * {{arg_dst}}_h);
+}
+//------------------ END KERNEL {{meta_kernel_id}} ---------------------
+)_";
+
+ return code;
+}
+
+void ClTemplateResize::declare_variables(GpuKernelVariableTable &vtable, const IGpuTemplateComponentWriter::ComponentGroup &comp_group) const
+{
+ vtable.declare_variable(
+ _src,
+ GpuKernelArgumentInfo(GpuKernelArgumentInfo::Type::Tensor_4D_t_Buffer),
+ comp_group.is_intermediate_tensor(_src),
+ "src");
+
+ vtable.declare_variable(
+ _dst,
+ GpuKernelArgumentInfo(GpuKernelArgumentInfo::Type::Tensor_4D_t_Buffer),
+ comp_group.is_intermediate_tensor(_dst),
+ "dst");
+}
+
+TagLUT ClTemplateResize::get_tag_lut(const GpuKernelVariableTable &vtable, const IGpuTemplateComponentWriter::ComponentGroup &comp_group) const
+{
+ TagLUT lut{};
+
+ // Arguments and global shared variables
+ lut["src"] = vtable.get_variable(_src);
+ lut["dst"] = vtable.get_variable(_dst);
+
+ const auto dst_argument = vtable.get_variable(comp_group.get_any_dst_tensor());
+ lut["arg_dst"] = dst_argument.uniq_name;
+
+ // Local build options
+ lut["meta_kernel_id"] = id();
+ lut["SRC_DATA_TYPE"] = get_cl_type_from_data_type(_src->data_type());
+ lut["SRC_TENSOR_TYPE"] = "BUFFER";
+ lut["DST_DATA_TYPE"] = get_cl_type_from_data_type(_dst->data_type());
+ lut["CONSTANT_VALUE"] = string_from_pixel_value(0, _src->data_type());
+
+ const bool is_qasymm_bilinear = is_data_type_quantized_asymmetric(_src->data_type())
+ && _attributes.interpolation_policy() == InterpolationPolicy::BILINEAR;
+
+ if(is_qasymm_bilinear)
+ {
+ const UniformQuantizationInfo qinfo = _src->quantization_info().uniform();
+ lut["SCALE"] = support::cpp11::to_string(qinfo.scale);
+ lut["OFFSET"] = support::cpp11::to_string(qinfo.offset);
+ }
+ else
+ {
+ lut["SCALE"] = support::cpp11::to_string(1);
+ lut["OFFSET"] = support::cpp11::to_string(0);
+ }
+
+ return lut;
+}
+
+CLBuildOptions ClTemplateResize::get_build_options(const IGpuTemplateComponentWriter::ComponentGroup &comp_group) const
+{
+ const Window root_window = comp_group.get_root_component()->template_writer()->get_window();
+ const unsigned int n0 = root_window.x().step();
+ const unsigned int m0 = root_window.y().step();
+ const unsigned int partial_n0 = _dst->dimension(0) % n0;
+
+ const float scale_x = scale_utils::calculate_resize_ratio(_src->dimension(1), _dst->dimension(1), _attributes.align_corners());
+ const float scale_y = scale_utils::calculate_resize_ratio(_src->dimension(2), _dst->dimension(2), _attributes.align_corners());
+
+ CLBuildOptions build_opts;
+
+ build_opts.add_option("-DN0=" + support::cpp11::to_string(n0));
+ build_opts.add_option("-DM0=" + support::cpp11::to_string(m0));
+ build_opts.add_option("-DPARTIAL_N0=" + support::cpp11::to_string(partial_n0));
+ build_opts.add_option("-DSCALE_X=" + float_to_string_with_full_precision(scale_x));
+ build_opts.add_option("-DSCALE_Y=" + float_to_string_with_full_precision(scale_y));
+
+ return build_opts;
+}
+
+std::string ClTemplateResize::get_config_id() const
+{
+ std::string config_id{};
+
+ config_id += "resize_";
+ config_id += (_attributes.interpolation_policy() == InterpolationPolicy::NEAREST_NEIGHBOR ? "NEAREST_NEIGHBOR" : "");
+ config_id += (_attributes.interpolation_policy() == InterpolationPolicy::BILINEAR ? "BILINEAR" : "");
+ config_id += "_";
+ config_id += (_attributes.sampling_policy() == SamplingPolicy::CENTER ? "center" : "topleft");
+ config_id += "_";
+ config_id += support::cpp11::to_string(_dst->dimension(0));
+ config_id += "_";
+ config_id += support::cpp11::to_string(_dst->dimension(1));
+ config_id += "_";
+ config_id += support::cpp11::to_string(_dst->dimension(2));
+ config_id += "_";
+ config_id += support::cpp11::to_string(_dst->dimension(3));
+
+ return config_id;
+}
+
+std::set<std::string> ClTemplateResize::get_headers_list() const
+{
+ return std::set<std::string>{ "helpers.h", "tile_helpers.h" };
+}
+
+Window ClTemplateResize::get_window() const
+{
+ ARM_COMPUTE_ERROR_ON_MSG(_dst->tensor_shape().total_size() == 0U, "Destination tensor is not initialized");
+
+ const unsigned int n0 = adjust_vec_size(16 / _src->element_size(), _src->dimension(0));
+ Window win = calculate_max_window(*_dst, Steps(n0));
+ return win.collapse(win, Window::DimZ);
+}
+
+} // namespace dynamic_fusion
+} // namespace experimental
+} // namespace arm_compute