// 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 #include #include #include #include class GAvgPoolCWMicrokernelTester { public: enum class Variant { Native, Scalar, }; inline GAvgPoolCWMicrokernelTester& elements(size_t elements) { assert(elements != 0); this->elements_ = elements; return *this; } inline size_t elements() const { return this->elements_; } inline GAvgPoolCWMicrokernelTester& channels(size_t channels) { assert(channels != 0); this->channels_ = channels; return *this; } inline size_t channels() const { return this->channels_; } inline GAvgPoolCWMicrokernelTester& qmin(uint8_t qmin) { this->qmin_ = qmin; return *this; } inline uint8_t qmin() const { return this->qmin_; } inline GAvgPoolCWMicrokernelTester& qmax(uint8_t qmax) { this->qmax_ = qmax; return *this; } inline uint8_t qmax() const { return this->qmax_; } inline GAvgPoolCWMicrokernelTester& iterations(size_t iterations) { this->iterations_ = iterations; return *this; } inline size_t iterations() const { return this->iterations_; } void Test(xnn_f32_gavgpool_cw_ukernel_function gavgpool, Variant variant = Variant::Native) const { std::random_device random_device; auto rng = std::mt19937(random_device()); std::uniform_real_distribution f32dist; std::vector x(elements() * channels() + XNN_EXTRA_BYTES / sizeof(float)); std::vector y(channels()); std::vector y_ref(channels()); for (size_t iteration = 0; iteration < iterations(); iteration++) { std::generate(x.begin(), x.end(), [&]() { return f32dist(rng); }); std::fill(y.begin(), y.end(), std::nanf("")); // Compute reference results, without clamping. for (size_t i = 0; i < channels(); i++) { float acc = 0.0f; for (size_t j = 0; j < elements(); j++) { acc += x[i * elements() + j]; } y_ref[i] = acc / float(elements()); } // Compute clamping parameters. 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 = accumulated_min + float(qmin()) / 255.0f * accumulated_range; const float y_max = accumulated_max - float(255 - qmax()) / 255.0f * accumulated_range; // Prepare parameters. union xnn_f32_gavgpool_params params; switch (variant) { case Variant::Native: xnn_init_f32_gavgpool_params( ¶ms, 1.0f / float(elements()), y_min, y_max, elements()); break; case Variant::Scalar: xnn_init_scalar_f32_gavgpool_params( ¶ms, 1.0f / float(elements()), y_min, y_max, elements()); break; } // Clamp reference results. for (float& y_value : y_ref) { y_value = std::max(std::min(y_value, y_max), y_min); } // Call optimized micro-kernel. gavgpool(elements() * sizeof(float), channels(), x.data(), y.data(), ¶ms); // Verify results. for (size_t i = 0; i < channels(); i++) { ASSERT_LE(y[i], y_max) << "at position " << i << ", elements = " << elements() << ", channels = " << channels(); ASSERT_GE(y[i], y_min) << "at position " << i << ", elements = " << elements() << ", channels = " << channels(); ASSERT_NEAR(y[i], y_ref[i], std::abs(y_ref[i]) * 1.0e-6f) << "at position " << i << ", elements = " << elements() << ", channels = " << channels(); } } } void Test(xnn_f16_gavgpool_cw_ukernel_function gavgpool, xnn_init_f16_gavgpool_neonfp16arith_params_fn init_params) const { std::random_device random_device; auto rng = std::mt19937(random_device()); std::uniform_real_distribution f32dist(0.1f, 10.0f); std::vector x(elements() * channels() + XNN_EXTRA_BYTES / sizeof(uint16_t)); std::vector y(channels()); std::vector y_ref(channels()); for (size_t iteration = 0; iteration < iterations(); iteration++) { std::generate(x.begin(), x.end(), [&]() { return fp16_ieee_from_fp32_value(f32dist(rng)); }); std::fill(y.begin(), y.end(), UINT16_C(0x7E00) /* NaN */); // Compute reference results, without clamping. for (size_t i = 0; i < channels(); i++) { float acc = 0.0f; for (size_t j = 0; j < elements(); j++) { acc += fp16_ieee_to_fp32_value(x[i * elements() + j]); } y_ref[i] = acc / float(elements()); } // Compute clamping parameters. 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 = fp16_ieee_to_fp32_value(fp16_ieee_from_fp32_value(accumulated_min + accumulated_range / 255.0f * float(qmin()))); const float y_max = fp16_ieee_to_fp32_value(fp16_ieee_from_fp32_value(accumulated_max - accumulated_range / 255.0f * float(255 - qmax()))); // Prepare parameters. union xnn_f16_gavgpool_params params; init_params( ¶ms, fp16_ieee_from_fp32_value(1.0f / float(elements())), fp16_ieee_from_fp32_value(y_min), fp16_ieee_from_fp32_value(y_max), elements()); // Clamp reference results. for (float& y_value : y_ref) { y_value = std::max(std::min(y_value, y_max), y_min); } // Call optimized micro-kernel. gavgpool(elements() * sizeof(uint16_t), channels(), x.data(), y.data(), ¶ms); // Verify results. for (size_t i = 0; i < channels(); i++) { ASSERT_LE(fp16_ieee_to_fp32_value(y[i]), y_max) << "at position " << i << ", elements = " << elements() << ", channels = " << channels(); ASSERT_GE(fp16_ieee_to_fp32_value(y[i]), y_min) << "at position " << i << ", elements = " << elements() << ", channels = " << channels(); ASSERT_NEAR(fp16_ieee_to_fp32_value(y[i]), y_ref[i], 1.0e-2f * std::abs(y_ref[i])) << "at position " << i << ", elements = " << elements() << ", channels = " << channels(); } } } private: size_t elements_{1}; size_t channels_{1}; uint8_t qmin_{0}; uint8_t qmax_{255}; size_t iterations_{15}; };