// Copyright (c) Facebook, Inc. and its affiliates. // All rights reserved. // // Copyright 2019 Google LLC // // This source code is licensed under the BSD-style license found in the // LICENSE file in the root directory of this source tree. #pragma once #include #include #include #include #include #include #include #include #include class AddOperatorTester { public: inline AddOperatorTester& channels(size_t channels) { assert(channels != 0); this->channels_ = channels; return *this; } inline size_t channels() const { return this->channels_; } inline AddOperatorTester& a_stride(size_t a_stride) { assert(a_stride != 0); this->a_stride_ = a_stride; return *this; } inline size_t a_stride() const { if (this->a_stride_ == 0) { return this->channels_; } else { assert(this->a_stride_ >= this->channels_); return this->a_stride_; } } inline AddOperatorTester& b_stride(size_t b_stride) { assert(b_stride != 0); this->b_stride_ = b_stride; return *this; } inline size_t b_stride() const { if (this->b_stride_ == 0) { return this->channels_; } else { assert(this->b_stride_ >= this->channels_); return this->b_stride_; } } inline AddOperatorTester& y_stride(size_t y_stride) { assert(y_stride != 0); this->y_stride_ = y_stride; return *this; } inline size_t y_stride() const { if (this->y_stride_ == 0) { return this->channels_; } else { assert(this->y_stride_ >= this->channels_); return this->y_stride_; } } inline AddOperatorTester& batch_size(size_t batch_size) { assert(batch_size != 0); this->batch_size_ = batch_size; return *this; } inline size_t batch_size() const { return this->batch_size_; } inline AddOperatorTester& a_scale(float a_scale) { assert(a_scale > 0.0f); assert(std::isnormal(a_scale)); this->a_scale_ = a_scale; return *this; } inline float a_scale() const { return this->a_scale_; } inline AddOperatorTester& a_zero_point(uint8_t a_zero_point) { this->a_zero_point_ = a_zero_point; return *this; } inline uint8_t a_zero_point() const { return this->a_zero_point_; } inline AddOperatorTester& b_scale(float b_scale) { assert(b_scale > 0.0f); assert(std::isnormal(b_scale)); this->b_scale_ = b_scale; return *this; } inline float b_scale() const { return this->b_scale_; } inline AddOperatorTester& b_zero_point(uint8_t b_zero_point) { this->b_zero_point_ = b_zero_point; return *this; } inline uint8_t b_zero_point() const { return this->b_zero_point_; } inline AddOperatorTester& y_scale(float y_scale) { assert(y_scale > 0.0f); assert(std::isnormal(y_scale)); this->y_scale_ = y_scale; return *this; } inline float y_scale() const { return this->y_scale_; } inline AddOperatorTester& y_zero_point(uint8_t y_zero_point) { this->y_zero_point_ = y_zero_point; return *this; } inline uint8_t y_zero_point() const { return this->y_zero_point_; } inline AddOperatorTester& qmin(uint8_t qmin) { this->qmin_ = qmin; return *this; } inline uint8_t qmin() const { return this->qmin_; } inline AddOperatorTester& qmax(uint8_t qmax) { this->qmax_ = qmax; return *this; } inline uint8_t qmax() const { return this->qmax_; } inline AddOperatorTester& iterations(size_t iterations) { this->iterations_ = iterations; return *this; } inline size_t iterations() const { return this->iterations_; } void TestQ8() const { std::random_device random_device; auto rng = std::mt19937(random_device()); auto u8rng = std::bind(std::uniform_int_distribution(), rng); std::vector a(XNN_EXTRA_BYTES / sizeof(uint8_t) + (batch_size() - 1) * a_stride() + channels()); std::vector b(XNN_EXTRA_BYTES / sizeof(uint8_t) + (batch_size() - 1) * b_stride() + channels()); std::vector y((batch_size() - 1) * y_stride() + channels()); std::vector y_ref(batch_size() * channels()); for (size_t iteration = 0; iteration < iterations(); iteration++) { std::generate(a.begin(), a.end(), std::ref(u8rng)); std::generate(b.begin(), b.end(), std::ref(u8rng)); std::fill(y.begin(), y.end(), 0xA5); if (batch_size() * channels() > 3) { ASSERT_NE(*std::max_element(a.cbegin(), a.cend()), *std::min_element(a.cbegin(), a.cend())); ASSERT_NE(*std::max_element(b.cbegin(), b.cend()), *std::min_element(b.cbegin(), b.cend())); } // Compute reference results. for (size_t i = 0; i < batch_size(); i++) { for (size_t c = 0; c < channels(); c++) { y_ref[i * channels() + c] = float(y_zero_point()) + float(int32_t(a[i * a_stride() + c]) - int32_t(a_zero_point())) * (a_scale() / y_scale()) + float(int32_t(b[i * b_stride() + c]) - int32_t(b_zero_point())) * (b_scale() / y_scale()); y_ref[i * channels() + c] = std::min(y_ref[i * channels() + c], float(qmax())); y_ref[i * channels() + c] = std::max(y_ref[i * channels() + c], float(qmin())); } } // Create, setup, run, and destroy Add operator. ASSERT_EQ(xnn_status_success, xnn_initialize(nullptr /* allocator */)); xnn_operator_t add_op = nullptr; ASSERT_EQ(xnn_status_success, xnn_create_add_nc_q8( channels(), a_stride(), b_stride(), y_stride(), a_zero_point(), a_scale(), b_zero_point(), b_scale(), y_zero_point(), y_scale(), qmin(), qmax(), 0, &add_op)); ASSERT_NE(nullptr, add_op); // Smart pointer to automatically delete add_op. std::unique_ptr auto_add_op(add_op, xnn_delete_operator); ASSERT_EQ(xnn_status_success, xnn_setup_add_nc_q8( add_op, batch_size(), a.data(), b.data(), y.data(), nullptr /* thread pool */)); ASSERT_EQ(xnn_status_success, xnn_run_operator(add_op, nullptr /* thread pool */)); /// Verify results. for (size_t i = 0; i < batch_size(); i++) { for (size_t c = 0; c < channels(); c++) { ASSERT_LE(uint32_t(y[i * y_stride() + c]), uint32_t(qmax())); ASSERT_GE(uint32_t(y[i * y_stride() + c]), uint32_t(qmin())); ASSERT_NEAR(float(int32_t(y[i * y_stride() + c])), y_ref[i * channels() + c], 0.6f); } } } } void TestF32() const { std::random_device random_device; auto rng = std::mt19937(random_device()); auto f32rng = std::bind(std::uniform_real_distribution(0.0f, 1.0f), rng); std::vector a(XNN_EXTRA_BYTES / sizeof(float) + (batch_size() - 1) * a_stride() + channels()); std::vector b(XNN_EXTRA_BYTES / sizeof(float) + (batch_size() - 1) * b_stride() + channels()); std::vector y((batch_size() - 1) * y_stride() + channels()); std::vector y_ref(batch_size() * channels()); for (size_t iteration = 0; iteration < iterations(); iteration++) { std::generate(a.begin(), a.end(), std::ref(f32rng)); std::generate(b.begin(), b.end(), std::ref(f32rng)); std::fill(y.begin(), y.end(), nanf("")); // Compute reference results. for (size_t i = 0; i < batch_size(); i++) { for (size_t c = 0; c < channels(); c++) { y_ref[i * channels() + c] = a[i * a_stride() + c] + b[i * b_stride() + c]; } } const float accumulated_min = *std::min_element(y_ref.cbegin(), y_ref.cend()); const float accumulated_max = *std::max_element(y_ref.cbegin(), y_ref.cend()); const float accumulated_range = accumulated_max - accumulated_min; const float y_min = batch_size() * channels() == 1 ? -std::numeric_limits::infinity() : accumulated_min + accumulated_range / 255.0f * float(qmin()); const float y_max = batch_size() * channels() == 1 ? +std::numeric_limits::infinity() : accumulated_max - accumulated_range / 255.0f * float(255 - qmax()); for (size_t i = 0; i < batch_size(); i++) { for (size_t c = 0; c < channels(); c++) { y_ref[i * channels() + c] = std::min(std::max(y_ref[i * channels() + c], y_min), y_max); } } // Create, setup, run, and destroy Add operator. ASSERT_EQ(xnn_status_success, xnn_initialize(nullptr /* allocator */)); xnn_operator_t add_op = nullptr; ASSERT_EQ(xnn_status_success, xnn_create_add_nc_f32( channels(), a_stride(), b_stride(), y_stride(), y_min, y_max, 0, &add_op)); ASSERT_NE(nullptr, add_op); // Smart pointer to automatically delete add_op. std::unique_ptr auto_add_op(add_op, xnn_delete_operator); ASSERT_EQ(xnn_status_success, xnn_setup_add_nc_f32( add_op, batch_size(), a.data(), b.data(), y.data(), nullptr /* thread pool */)); ASSERT_EQ(xnn_status_success, xnn_run_operator(add_op, nullptr /* thread pool */)); // Verify results. for (size_t i = 0; i < batch_size(); i++) { for (size_t c = 0; c < channels(); c++) { ASSERT_NEAR(y[i * y_stride() + c], y_ref[i * channels() + c], 1.0e-6f * y_ref[i * channels() + c]) << "i = " << i << ", c = " << c; } } } } private: size_t batch_size_{1}; size_t channels_{1}; size_t a_stride_{0}; size_t b_stride_{0}; size_t y_stride_{0}; float a_scale_{0.75f}; float b_scale_{1.25f}; float y_scale_{0.96875f}; uint8_t a_zero_point_{121}; uint8_t b_zero_point_{127}; uint8_t y_zero_point_{133}; uint8_t qmin_{0}; uint8_t qmax_{255}; size_t iterations_{15}; };