wehub-resource-sync
1418 lines
60 KiB
C++
1418 lines
60 KiB
C++
/* Copyright 2023 The TensorFlow Authors. All Rights Reserved.
|
|
|
|
Licensed under the Apache License, Version 2.0 (the "License");
|
|
you may not use this file except in compliance with the License.
|
|
You may obtain a copy of the License at
|
|
|
|
http://www.apache.org/licenses/LICENSE-2.0
|
|
|
|
Unless required by applicable law or agreed to in writing, software
|
|
distributed under the License is distributed on an "AS IS" BASIS,
|
|
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
See the License for the specific language governing permissions and
|
|
limitations under the License.
|
|
==============================================================================*/
|
|
#include <stddef.h>
|
|
#include <stdint.h>
|
|
|
|
#include <algorithm>
|
|
#include <array>
|
|
#include <complex>
|
|
#include <functional>
|
|
#include <numeric>
|
|
#include <random>
|
|
#include <utility>
|
|
#include <vector>
|
|
|
|
#include <gmock/gmock.h>
|
|
#include <gtest/gtest.h>
|
|
#include "flatbuffers/flatbuffers.h" // from @flatbuffers
|
|
#include "tensorflow/lite/kernels/test_util.h"
|
|
#include "tensorflow/lite/schema/schema_generated.h"
|
|
|
|
namespace tflite {
|
|
namespace {
|
|
|
|
using ::testing::ElementsAreArray;
|
|
|
|
template <typename InputType, typename QuantizedType = InputType>
|
|
class BaseMulOpModel : public SingleOpModel {
|
|
public:
|
|
BaseMulOpModel(const TensorData& input1, const TensorData& input2,
|
|
const TensorData& output,
|
|
ActivationFunctionType activation_type,
|
|
const std::vector<InputType>& input1_data,
|
|
const std::vector<InputType>& input2_data,
|
|
bool constant_tensors) {
|
|
if (constant_tensors) {
|
|
input1_ = AddConstInput(input1, input1_data);
|
|
input2_ = AddConstInput(input2, input2_data);
|
|
} else {
|
|
input1_ = AddInput(input1);
|
|
input2_ = AddInput(input2);
|
|
}
|
|
output_ = AddOutput(output);
|
|
SetBuiltinOp(BuiltinOperator_MUL, BuiltinOptions_MulOptions,
|
|
CreateMulOptions(builder_, activation_type).Union());
|
|
SetBypassDefaultDelegates();
|
|
BuildInterpreter({GetShape(input1_), GetShape(input2_)});
|
|
if (!constant_tensors) {
|
|
PopulateTensor<QuantizedType>(input1_, input1_data);
|
|
PopulateTensor<QuantizedType>(input2_, input2_data);
|
|
}
|
|
}
|
|
|
|
int input1() { return input1_; }
|
|
int input2() { return input2_; }
|
|
|
|
void Resize(const std::vector<int>& input1_shape,
|
|
const std::vector<int>& input2_shape) {
|
|
interpreter_->ResizeInputTensor(input1_, input1_shape);
|
|
interpreter_->ResizeInputTensor(input2_, input2_shape);
|
|
AllocateTensors();
|
|
}
|
|
|
|
std::vector<InputType> GetOutput() {
|
|
return ExtractVector<InputType>(output_);
|
|
}
|
|
|
|
protected:
|
|
int input1_;
|
|
int input2_;
|
|
int output_;
|
|
};
|
|
|
|
template <typename T>
|
|
class MulOpModel : public BaseMulOpModel<T> {
|
|
public:
|
|
using BaseMulOpModel<T>::BaseMulOpModel;
|
|
};
|
|
|
|
template <typename T>
|
|
class FloatMulTest : public ::testing::Test {};
|
|
|
|
using FloatMulTestTypes = ::testing::Types<float, half, Eigen::bfloat16>;
|
|
TYPED_TEST_SUITE(FloatMulTest, FloatMulTestTypes);
|
|
|
|
class ComplexMulOpModel : public BaseMulOpModel<std::complex<float>> {
|
|
public:
|
|
using BaseMulOpModel::BaseMulOpModel;
|
|
};
|
|
|
|
template <typename InputType>
|
|
class IntegerMulOpModel : public BaseMulOpModel<InputType> {
|
|
public:
|
|
using BaseMulOpModel<InputType>::BaseMulOpModel;
|
|
};
|
|
|
|
// For quantized Mul, the error shouldn't exceed (2*step + step^2).
|
|
// The param min=-1.0 & max=1.0 is used in the following tests.
|
|
// The tolerance value is ~0.0157.
|
|
const float kQuantizedStep = 2.0 / 255.0;
|
|
const float kQuantizedTolerance =
|
|
2.0 * kQuantizedStep + kQuantizedStep * kQuantizedStep;
|
|
const float kQuantizedStepInt16 = 2.0 / 32767.0;
|
|
const float kQuantizedToleranceInt16 =
|
|
2.0 * kQuantizedStepInt16 + kQuantizedStepInt16 * kQuantizedStepInt16;
|
|
|
|
template <typename InputType, typename QuantizedType>
|
|
class QuantizedMulOpModel : public SingleOpModel {
|
|
public:
|
|
QuantizedMulOpModel(const TensorData& input1, const TensorData& input2,
|
|
const TensorData& output,
|
|
ActivationFunctionType activation_type,
|
|
const std::vector<float>& input1_data,
|
|
const std::vector<float>& input2_data,
|
|
bool constant_tensors) {
|
|
if (constant_tensors) {
|
|
std::vector<InputType> quantized_input1_data(input1_data.size());
|
|
std::vector<InputType> quantized_input2_data(input2_data.size());
|
|
std::pair<float, int32_t> input1_quantization_params =
|
|
QuantizationParams<InputType>(input1.min, input1.max);
|
|
std::pair<float, int32_t> input2_quantization_params =
|
|
QuantizationParams<InputType>(input2.min, input2.max);
|
|
quantized_input1_data =
|
|
Quantize<InputType>(input1_data, input1_quantization_params.first,
|
|
input1_quantization_params.second);
|
|
quantized_input2_data =
|
|
Quantize<InputType>(input2_data, input2_quantization_params.first,
|
|
input2_quantization_params.second);
|
|
input1_ = AddConstInput(input1, quantized_input1_data);
|
|
input2_ = AddConstInput(input2, quantized_input2_data);
|
|
} else {
|
|
input1_ = AddInput(input1);
|
|
input2_ = AddInput(input2);
|
|
}
|
|
output_ = AddOutput(output);
|
|
SetBuiltinOp(BuiltinOperator_MUL, BuiltinOptions_MulOptions,
|
|
CreateMulOptions(builder_, activation_type).Union());
|
|
BuildInterpreter({GetShape(input1_), GetShape(input2_)});
|
|
if (!constant_tensors) {
|
|
QuantizeAndPopulate<InputType>(input1_, input1_data);
|
|
QuantizeAndPopulate<InputType>(input2_, input2_data);
|
|
}
|
|
}
|
|
|
|
int input1() { return input1_; }
|
|
int input2() { return input2_; }
|
|
|
|
void Resize(const std::vector<int>& input1_shape,
|
|
const std::vector<int>& input2_shape) {
|
|
interpreter_->ResizeInputTensor(input1_, input1_shape);
|
|
interpreter_->ResizeInputTensor(input2_, input2_shape);
|
|
AllocateTensors();
|
|
}
|
|
|
|
std::vector<float> GetDequantizedOutput() {
|
|
return Dequantize<QuantizedType>(
|
|
this->template ExtractVector<QuantizedType>(this->output_),
|
|
GetScale(this->output_), GetZeroPoint(this->output_));
|
|
}
|
|
|
|
protected:
|
|
int input1_;
|
|
int input2_;
|
|
int output_;
|
|
};
|
|
|
|
using MulOpTest = testing::TestWithParam<bool>;
|
|
|
|
TYPED_TEST(FloatMulTest, NoActivationInplaceInput0) {
|
|
using T = TypeParam;
|
|
MulOpModel<T> m({GetTensorType<T>(), {1, 2, 2, 1}},
|
|
{GetTensorType<T>(), {1, 2, 2, 1}}, {GetTensorType<T>(), {}},
|
|
ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8}),
|
|
ToVector<T>({0.1, 0.2, 0.3, 0.5}), false);
|
|
const int kInplaceInputTensorIdx = 0;
|
|
const int kInplaceOutputTensorIdx = 0;
|
|
const TfLiteTensor* input_tensor = m.GetInputTensor(kInplaceInputTensorIdx);
|
|
TfLiteTensor* output_tensor = m.GetOutputTensor(kInplaceOutputTensorIdx);
|
|
output_tensor->data.data = input_tensor->data.data;
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.21, 0.4},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)));
|
|
EXPECT_EQ(output_tensor->data.data, input_tensor->data.data);
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, NoActivationInplaceInput1) {
|
|
using T = TypeParam;
|
|
MulOpModel<T> m({GetTensorType<T>(), {1, 2, 2, 1}},
|
|
{GetTensorType<T>(), {1, 2, 2, 1}}, {GetTensorType<T>(), {}},
|
|
ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8}),
|
|
ToVector<T>({0.1, 0.2, 0.3, 0.5}), false);
|
|
const int kInplaceInputTensorIdx = 1;
|
|
const int kInplaceOutputTensorIdx = 0;
|
|
const TfLiteTensor* input_tensor = m.GetInputTensor(kInplaceInputTensorIdx);
|
|
TfLiteTensor* output_tensor = m.GetOutputTensor(kInplaceOutputTensorIdx);
|
|
output_tensor->data.data = input_tensor->data.data;
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.21, 0.4},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)));
|
|
EXPECT_EQ(output_tensor->data.data, input_tensor->data.data);
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, NoActivation) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
MulOpModel<T> m({GetTensorType<T>(), {1, 2, 2, 1}},
|
|
{GetTensorType<T>(), {1, 2, 2, 1}},
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8}),
|
|
ToVector<T>({0.1, 0.2, 0.3, 0.5}), constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.21, 0.4},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "with constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, ActivationRELU_N1_TO_1) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
MulOpModel<T> m(
|
|
{GetTensorType<T>(), {1, 2, 2, 1}}, {GetTensorType<T>(), {1, 2, 2, 1}},
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_RELU_N1_TO_1,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8}), ToVector<T>({0.1, 0.2, 0.3, 5}),
|
|
constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.21, 1.0},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "with constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, VariousInputShapes) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
MulOpModel<T> m({GetTensorType<T>(), test_shapes[i]},
|
|
{GetTensorType<T>(), test_shapes[i]},
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8, 1.1, 2.0}),
|
|
ToVector<T>({0.1, 0.2, 0.3, 0.5, 1.1, 0.1}),
|
|
constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.21, 0.4, 1.21, 0.2},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "With shape number " << i
|
|
<< " and constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, WithScalarBroadcast) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
MulOpModel<T> m({GetTensorType<T>(), test_shapes[i]},
|
|
{GetTensorType<T>(), {}}, // always a scalar
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8, 1.1, 2.0}),
|
|
ToVector<T>({0.1}), constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.02, 0.07, 0.08, 0.11, 0.2},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "With shape number " << i
|
|
<< " and constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, WithBroadcast) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{2, 4}, {2, 1, 4}, {1, 2, 4}, {1, 2, 1, 4}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
MulOpModel<T> m({GetTensorType<T>(), test_shapes[i]},
|
|
{GetTensorType<T>(), {4}}, // always a scalar
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8, 1.1, 2.0, 1.1, 0.8}),
|
|
ToVector<T>({0.1, 0.2, 0.3, 0.4}), constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.21, 0.32, 0.11, 0.4, 0.33, 0.32},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "With shape number " << i
|
|
<< " and constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, MixedBroadcast) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
const std::vector<int> base_shape = {2, 3, 1, 2};
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{1, 1, 3, 2}, {1, 3, 1, 2}, {2, 1, 3, 1}, {2, 3, 1, 1}};
|
|
const std::vector<std::vector<float>> test_outputs = {
|
|
{-0.06f, 0.69f, 0.12f, 1.15f, -0.30f, 2.07f, 0.18f, 0.15f, -0.36f,
|
|
0.25f, 0.90f, 0.45f, 0.16f, -0.33f, -0.32f, -0.55f, 0.80f, -0.99f,
|
|
0.24f, 0.84f, -0.48f, 1.40f, 1.20f, 2.52f, -0.32f, 0.00f, 0.64f,
|
|
0.00f, -1.60f, 0.00f, 0.14f, -0.66f, -0.28f, -1.10f, 0.70f, -1.98f},
|
|
{-0.06f, 0.69f, -0.36f, 0.25f, 0.80f, -0.99f, 0.24f, 0.84f, 0.64f,
|
|
0.00f, 0.70f, -1.98f},
|
|
{-0.06f, 0.46f, -0.09f, 0.69f, 0.12f, -0.92f, 0.18f, 0.10f, 0.27f,
|
|
0.15f, -0.36f, -0.20f, 0.16f, -0.22f, 0.24f, -0.33f, -0.32f, 0.44f,
|
|
0.60f, 1.40f, 1.20f, 2.80f, 1.08f, 2.52f, -0.80f, 0.00f, -1.60f,
|
|
0.00f, -1.44f, 0.00f, 0.35f, -1.10f, 0.70f, -2.20f, 0.63f, -1.98f},
|
|
{-0.06f, 0.46f, 0.27f, 0.15f, -0.32f, 0.44f, 0.60f, 1.40f, -1.60f,
|
|
0.00f, 0.63f, -1.98f}};
|
|
for (size_t i = 0; i < test_shapes.size(); ++i) {
|
|
MulOpModel<T> model_fixture(
|
|
{GetTensorType<T>(), base_shape},
|
|
{GetTensorType<T>(), test_shapes[i]}, {GetTensorType<T>(), {}},
|
|
ActivationFunctionType_NONE,
|
|
ToVector<T>({-0.3f, 2.3f, 0.9f, 0.5f, 0.8f, -1.1f, 1.2f, 2.8f, -1.6f,
|
|
0.0f, 0.7f, -2.2f}),
|
|
ToVector<T>({0.2f, 0.3f, -0.4f, 0.5f, 1.0f, 0.9f}), constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &model_fixture);
|
|
|
|
EXPECT_THAT(model_fixture.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
test_outputs[i],
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "With shape number " << i
|
|
<< " and constant_tensors=" << constant_tensors;
|
|
}
|
|
// Re-run with exchanged inputs.
|
|
for (size_t i = 0; i < test_shapes.size(); ++i) {
|
|
MulOpModel<T> model_fixture(
|
|
{GetTensorType<T>(), test_shapes[i]},
|
|
{GetTensorType<T>(), base_shape}, {GetTensorType<T>(), {}},
|
|
ActivationFunctionType_NONE,
|
|
ToVector<T>({0.2f, 0.3f, -0.4f, 0.5f, 1.0f, 0.9f}),
|
|
ToVector<T>({-0.3f, 2.3f, 0.9f, 0.5f, 0.8f, -1.1f, 1.2f, 2.8f, -1.6f,
|
|
0.0f, 0.7f, -2.2f}),
|
|
constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &model_fixture);
|
|
EXPECT_THAT(model_fixture.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
test_outputs[i],
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "With shape number " << i
|
|
<< " and constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, WithBroadcast2Elements) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{2, 2}, {2, 1, 2}, {1, 2, 2}, {1, 2, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
MulOpModel<T> m({GetTensorType<T>(), test_shapes[i]},
|
|
{GetTensorType<T>(), {2}}, // always a scalar
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE,
|
|
ToVector<T>({-2.0, 0.2, 0.7, 0.8}),
|
|
ToVector<T>({0.1, 0.2}), constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{-0.2, 0.04, 0.07, 0.16},
|
|
static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "With shape number " << i
|
|
<< " and constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
}
|
|
|
|
TYPED_TEST(FloatMulTest, ScalarAndOneElement) {
|
|
using T = TypeParam;
|
|
for (bool constant_tensors : {false, true}) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
continue;
|
|
}
|
|
MulOpModel<T> m({GetTensorType<T>(), {1}}, {GetTensorType<T>(), {}},
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE,
|
|
ToVector<T>({0.8}), ToVector<T>({0.5}), constant_tensors);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(
|
|
m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear(
|
|
{0.4}, static_cast<float>(NumericLimits<T>::epsilon()) * 10)))
|
|
<< "with constant_tensors=" << constant_tensors;
|
|
}
|
|
}
|
|
|
|
TEST_P(MulOpTest, IntegerNoActivation) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
IntegerMulOpModel<int32_t> m(
|
|
{TensorType_INT32, {1, 2, 2, 1}}, {TensorType_INT32, {1, 2, 2, 1}},
|
|
{TensorType_INT32, {}}, ActivationFunctionType_NONE, {-20, 2, 7, 8},
|
|
{1, 2, 3, 5}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-20, 4, 21, 40}));
|
|
}
|
|
|
|
TEST_P(MulOpTest, Int16ActivationRELU_N1_TO_1) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
IntegerMulOpModel<int16_t> m(
|
|
{TensorType_INT16, {1, 2, 2, 1}}, {TensorType_INT16, {1, 2, 2, 1}},
|
|
{TensorType_INT16, {}}, ActivationFunctionType_RELU_N1_TO_1,
|
|
{-20, 2, 7, 8}, {1, 2, 3, 5}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-1, 1, 1, 1}));
|
|
}
|
|
|
|
TEST_P(MulOpTest, Int16VariousInputShapes) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
IntegerMulOpModel<int16_t> m(
|
|
{TensorType_INT16, test_shapes[i]}, {TensorType_INT16, test_shapes[i]},
|
|
{TensorType_INT16, {}}, ActivationFunctionType_NONE,
|
|
{-20, 2, 7, 8, 11, 20}, {1, 2, 3, 5, 11, 1}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-20, 4, 21, 40, 121, 20}))
|
|
<< "With shape number " << i;
|
|
}
|
|
}
|
|
|
|
TEST_P(MulOpTest, Int16WithBroadcast) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
IntegerMulOpModel<int16_t> m({TensorType_INT16, test_shapes[i]},
|
|
{TensorType_INT16, {}}, // always a scalar
|
|
{TensorType_INT16, {}},
|
|
ActivationFunctionType_NONE,
|
|
{-20, 2, 7, 8, 11, 20}, {1}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-20, 2, 7, 8, 11, 20}))
|
|
<< "With shape number " << i;
|
|
}
|
|
}
|
|
|
|
TEST_P(MulOpTest, 16BitIntegerNoActivation) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
IntegerMulOpModel<int16_t> m({TensorType_INT16, {4}}, {TensorType_INT16, {4}},
|
|
{TensorType_INT16, {}},
|
|
ActivationFunctionType_NONE, {-20, 2, 7, 8},
|
|
{1, 2, 3, 5}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-20, 4, 21, 40}));
|
|
}
|
|
|
|
TEST_P(MulOpTest, 32BitUnsignedIntegerNoActivation) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
IntegerMulOpModel<uint32_t> m(
|
|
{TensorType_UINT32, {1, 2, 2, 1}}, {TensorType_UINT32, {1, 2, 2, 1}},
|
|
{TensorType_UINT32, {}}, ActivationFunctionType_NONE, {20, 2, 7, 8},
|
|
{1, 2, 3, 5}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({20, 4, 21, 40}));
|
|
}
|
|
|
|
TEST_P(MulOpTest, ComplexBaseTest) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
ComplexMulOpModel m({TensorType_COMPLEX64, {1, 2, 2, 1}},
|
|
{TensorType_COMPLEX64, {1, 2, 2, 1}},
|
|
{TensorType_COMPLEX64, {}}, ActivationFunctionType_NONE,
|
|
{-20, {2, 3}, {7, 2}, 8}, {1, {2, -3}, {3, -4}, 5},
|
|
constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
std::complex<float> expected_result[4] = {-20, 13, {29, -22}, 40};
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray(expected_result));
|
|
}
|
|
|
|
TEST_P(MulOpTest, ComplexWithBroadcast) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
ComplexMulOpModel m({TensorType_COMPLEX64, test_shapes[i]},
|
|
{TensorType_COMPLEX64, {}}, {TensorType_COMPLEX64, {}},
|
|
ActivationFunctionType_NONE, {-20, 2, 7, 8, 11, 20},
|
|
{1}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-20, 2, 7, 8, 11, 20}))
|
|
<< "With shape number " << i;
|
|
}
|
|
}
|
|
|
|
#if GTEST_HAS_DEATH_TEST
|
|
TEST(MulOpTest, IncompatibleActivation) {
|
|
EXPECT_DEATH(ComplexMulOpModel({TensorType_COMPLEX64, {1, 2, 2, 1}},
|
|
{TensorType_COMPLEX64, {1, 2, 2, 1}},
|
|
{TensorType_COMPLEX64, {}},
|
|
ActivationFunctionType_RELU_N1_TO_1,
|
|
{1, 2, 3, 4}, {1, 2, 3, 4}, false),
|
|
"Activation is not allowed for COMPLEX64 input");
|
|
}
|
|
#endif
|
|
|
|
TEST_P(MulOpTest, Int32ActivationRELU_N1_TO_1) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
IntegerMulOpModel<int32_t> m(
|
|
{TensorType_INT32, {1, 2, 2, 1}}, {TensorType_INT32, {1, 2, 2, 1}},
|
|
{TensorType_INT32, {}}, ActivationFunctionType_RELU_N1_TO_1,
|
|
{-20, 2, 7, 8}, {1, 2, 3, 5}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-1, 1, 1, 1}));
|
|
}
|
|
|
|
TEST_P(MulOpTest, Int32VariousInputShapes) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
IntegerMulOpModel<int32_t> m(
|
|
{TensorType_INT32, test_shapes[i]}, {TensorType_INT32, test_shapes[i]},
|
|
{TensorType_INT32, {}}, ActivationFunctionType_NONE,
|
|
{-20, 2, 7, 8, 11, 20}, {1, 2, 3, 5, 11, 1}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), ElementsAreArray({-20, 4, 21, 40, 121, 20}))
|
|
<< "With shape number " << i;
|
|
}
|
|
}
|
|
|
|
// Neon intrinsics are only dispatched when tensor has at least 16 elements.
|
|
TEST_P(MulOpTest, Int32LargeInputShapeNoActivation) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<int> test_shape = {4, 4, 4, 4};
|
|
constexpr int kFlatSize = 4 * 4 * 4 * 4;
|
|
|
|
std::vector<int> lhs_data(kFlatSize);
|
|
std::iota(lhs_data.begin(), lhs_data.end(), 0);
|
|
|
|
std::vector<int> rhs_data(kFlatSize);
|
|
std::iota(rhs_data.begin(), rhs_data.end(), 0);
|
|
|
|
IntegerMulOpModel<int32_t> m(
|
|
{TensorType_INT32, test_shape}, {TensorType_INT32, test_shape},
|
|
{TensorType_INT32, {}}, ActivationFunctionType_NONE, lhs_data, rhs_data,
|
|
constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
|
|
const std::vector<int> output = m.GetOutput();
|
|
ASSERT_EQ(output.size(), kFlatSize);
|
|
for (int i = 0; i < kFlatSize; ++i) {
|
|
EXPECT_EQ(output[i], i * i);
|
|
}
|
|
}
|
|
|
|
// Neon intrinsics are only dispatched when tensor has at least 16 elements.
|
|
TEST_P(MulOpTest, Int32LargeInputShapeRELU6) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<int> test_shape = {4, 4, 4, 4};
|
|
constexpr int kFlatSize = 4 * 4 * 4 * 4;
|
|
|
|
std::vector<int> lhs_data(kFlatSize);
|
|
std::iota(lhs_data.begin(), lhs_data.end(), 0);
|
|
|
|
std::vector<int> rhs_data(kFlatSize);
|
|
std::iota(rhs_data.begin(), rhs_data.end(), 0);
|
|
|
|
IntegerMulOpModel<int32_t> m(
|
|
{TensorType_INT32, test_shape}, {TensorType_INT32, test_shape},
|
|
{TensorType_INT32, {}}, ActivationFunctionType_RELU6, lhs_data, rhs_data,
|
|
constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
|
|
const std::vector<int> output = m.GetOutput();
|
|
ASSERT_EQ(output.size(), kFlatSize);
|
|
for (int i = 0; i < kFlatSize; ++i) {
|
|
EXPECT_EQ(output[i], std::min(i * i, 6));
|
|
}
|
|
}
|
|
|
|
TEST_P(MulOpTest, Int32WithBroadcast) {
|
|
bool constant_tensors = GetParam();
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
IntegerMulOpModel<int32_t> m({TensorType_INT32, test_shapes[i]},
|
|
{TensorType_INT32, {}}, // always a scalar
|
|
{TensorType_INT32, {}},
|
|
ActivationFunctionType_NONE,
|
|
{-20, 2, 7, 8, 11, 20}, {1}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
ElementsAreArray(ArrayFloatNear({-20, 2, 7, 8, 11, 20})))
|
|
<< "With shape number " << i;
|
|
}
|
|
}
|
|
|
|
template <TensorType tensor_type, typename integer_dtype>
|
|
void NoActivation(bool constant_tensors) {
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
QuantizedMulOpModel<integer_dtype, integer_dtype> m(
|
|
{tensor_type, {1, 2, 2, 1}, -1.0, 1.0},
|
|
{tensor_type, {1, 2, 2, 1}, -1.0, 1.0}, {tensor_type, {}, -1.0, 1.0},
|
|
ActivationFunctionType_NONE, {-0.8, 0.2, 0.9, 0.7}, {0.6, 0.4, 0.9, 0.8},
|
|
constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear({-0.48, 0.08, 0.81, 0.56},
|
|
kQuantizedTolerance)));
|
|
}
|
|
|
|
template <TensorType tensor_type, typename integer_dtype>
|
|
void NoActivationLargeMultiplier(bool constant_tensors) {
|
|
// Intentionally pathological output range much narrower than needed
|
|
// to represent input values to exercise the multiplier>1 case.
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
QuantizedMulOpModel<integer_dtype, integer_dtype> m(
|
|
{tensor_type, {1, 2, 2, 1}, -100, 100},
|
|
{tensor_type, {1, 2, 2, 1}, -100, 100}, {tensor_type, {}, -10, 10},
|
|
ActivationFunctionType_NONE, {-4, 2, 3, 1}, {-1, -3, 4, 2},
|
|
constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
// Note the large tolerance. This computation is inherently inaccurate.
|
|
const float kTolerance = 1.4f;
|
|
EXPECT_THAT(m.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear({4, -6, 10, 2}, kTolerance)));
|
|
}
|
|
|
|
TEST_P(MulOpTest, NoActivationUInt8) {
|
|
bool constant_tensors = GetParam();
|
|
NoActivation<TensorType_UINT8, uint8_t>(constant_tensors);
|
|
NoActivationLargeMultiplier<TensorType_UINT8, uint8_t>(constant_tensors);
|
|
}
|
|
|
|
TEST_P(MulOpTest, NoActivationInt8) {
|
|
bool constant_tensors = GetParam();
|
|
NoActivation<TensorType_INT8, int8_t>(constant_tensors);
|
|
NoActivationLargeMultiplier<TensorType_INT8, int8_t>(constant_tensors);
|
|
}
|
|
|
|
TEST_P(MulOpTest, NoActivationInt16) {
|
|
bool constant_tensors = GetParam();
|
|
const float kMin = -1.f;
|
|
const float kMax = 32767.f / 32768.f;
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
QuantizedMulOpModel<int16_t, int16_t> m(
|
|
{TensorType_INT16, {1, 2, 2, 1}, kMin, kMax},
|
|
{TensorType_INT16, {1, 2, 2, 1}, kMin, kMax},
|
|
{TensorType_INT16, {}, kMin, kMax}, ActivationFunctionType_NONE,
|
|
{-0.8, 0.2, 0.9, 0.7}, {0.6, 0.4, 0.9, 0.8}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear({-0.48, 0.08, 0.81, 0.56},
|
|
kQuantizedToleranceInt16)));
|
|
}
|
|
|
|
TEST_P(MulOpTest, NoActivationInt16Scaled) {
|
|
bool constant_tensors = GetParam();
|
|
const float kMin = -2.f;
|
|
const float kMax = 2.f * 32767.f / 32768.f;
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
QuantizedMulOpModel<int16_t, int16_t> m(
|
|
{TensorType_INT16, {1, 2, 3, 1}, kMin, kMax},
|
|
{TensorType_INT16, {1, 2, 3, 1}, 2 * kMin, 2 * kMax},
|
|
{TensorType_INT16, {}, 8 * kMin, 8 * kMax}, ActivationFunctionType_NONE,
|
|
{-1.8, 0.2, 0.9, 1.7, 0.1, -1.95}, {3.6, -3.4, 3.9, 0.8, -1.0, -3.95},
|
|
constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
|
|
const float kQuantizedToleranceInt16Scaled =
|
|
6.0 * kQuantizedStepInt16 + kQuantizedStepInt16 * kQuantizedStepInt16;
|
|
|
|
EXPECT_THAT(
|
|
m.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear({-6.48, -0.68, 3.51, 1.36, -0.1, 7.7025},
|
|
kQuantizedToleranceInt16Scaled)));
|
|
}
|
|
|
|
template <TensorType tensor_type, typename integer_dtype>
|
|
void NoActivationInt16With8BitOutput(bool constant_tensors) {
|
|
const float kMinInt16 = -1.f;
|
|
const float kMaxInt16 = 32767.f / 32768.f;
|
|
const float kMinUint8 = -1.f;
|
|
const float kMaxUint8 = 127.f / 128.f;
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
QuantizedMulOpModel<int16_t, integer_dtype> m(
|
|
{TensorType_INT16, {1, 2, 2, 1}, kMinInt16, kMaxInt16},
|
|
{TensorType_INT16, {1, 2, 2, 1}, kMinInt16, kMaxInt16},
|
|
{tensor_type, {}, kMinUint8, kMaxUint8}, ActivationFunctionType_NONE,
|
|
{-0.8, 0.2, 0.9, 0.7}, {0.6, 0.4, 0.9, 0.8}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear({-0.48, 0.08, 0.81, 0.56},
|
|
kQuantizedTolerance)));
|
|
}
|
|
|
|
TEST_P(MulOpTest, NoActivationInt16WithUint8Output) {
|
|
bool constant_tensors = GetParam();
|
|
NoActivationInt16With8BitOutput<TensorType_UINT8, uint8_t>(constant_tensors);
|
|
}
|
|
|
|
TEST_P(MulOpTest, NoActivationInt16Withint8Output) {
|
|
bool constant_tensors = GetParam();
|
|
NoActivationInt16With8BitOutput<TensorType_INT8, int8_t>(constant_tensors);
|
|
}
|
|
|
|
// for quantized Mul, the error shouldn't exceed 2*step
|
|
float GetTolerance(int min, int max) {
|
|
float kQuantizedStep = (max - min) / 255.0;
|
|
float kQuantizedTolerance = 2.0 * kQuantizedStep;
|
|
return kQuantizedTolerance;
|
|
}
|
|
|
|
template <TensorType tensor_type, typename integer_dtype>
|
|
void WithBroadcast(bool constant_tensors) {
|
|
const float kQuantizedTolerance = GetTolerance(-3.0, 3.0);
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{6}, {2, 3}, {2, 1, 3}, {1, 3, 1, 2}};
|
|
// Test with a smaller than 1 and greater than 1 quantization multiplier
|
|
const std::vector<std::pair<float, float>> test_input_range = {{-3.0, 3.0},
|
|
{-6.0, 6.0}};
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
for (int i = 0; i < test_shapes.size(); ++i) {
|
|
for (int j = 0; j < test_input_range.size(); ++j) {
|
|
const std::pair<float, float>& input_range = test_input_range[j];
|
|
QuantizedMulOpModel<integer_dtype, integer_dtype> m(
|
|
{tensor_type, test_shapes[i], input_range.first, input_range.second},
|
|
{tensor_type, {}, input_range.first, input_range.second},
|
|
{tensor_type, {}, -0.2, 0.2}, ActivationFunctionType_NONE,
|
|
{-2.0, 0.2, 0.7, 0.8, 1.1, 2.0}, {0.1}, constant_tensors);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(
|
|
m.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear({-0.2, 0.02, 0.07, 0.08, 0.11, 0.2},
|
|
kQuantizedTolerance)))
|
|
<< "With shape number " << i << " and range number " << j;
|
|
}
|
|
}
|
|
}
|
|
|
|
template <enum TensorType tensor_type, typename integer_dtype>
|
|
void QuantizedWithMixedBroadcast(bool constant_tensors) {
|
|
const float kQuantizedTolerance = GetTolerance(-3.f, 3.f);
|
|
const std::vector<int> base_shape = {2, 3, 1, 2};
|
|
const std::vector<std::vector<int>> test_shapes = {
|
|
{1, 1, 3, 2}, {1, 3, 1, 2}, {2, 1, 3, 1}, {2, 3, 1, 1}};
|
|
const std::vector<std::vector<float>> test_outputs = {
|
|
{-0.06f, 0.69f, 0.12f, 1.15f, -0.30f, 2.07f, 0.18f, 0.15f, -0.36f,
|
|
0.25f, 0.90f, 0.45f, 0.16f, -0.33f, -0.32f, -0.55f, 0.80f, -0.99f,
|
|
0.24f, 0.84f, -0.48f, 1.40f, 1.20f, 2.52f, -0.32f, 0.00f, 0.64f,
|
|
0.00f, -1.60f, 0.00f, 0.14f, -0.66f, -0.28f, -1.10f, 0.70f, -1.98f},
|
|
{-0.06f, 0.69f, -0.36f, 0.25f, 0.80f, -0.99f, 0.24f, 0.84f, 0.64f, 0.00f,
|
|
0.70f, -1.98f},
|
|
{-0.06f, 0.46f, -0.09f, 0.69f, 0.12f, -0.92f, 0.18f, 0.10f, 0.27f,
|
|
0.15f, -0.36f, -0.20f, 0.16f, -0.22f, 0.24f, -0.33f, -0.32f, 0.44f,
|
|
0.60f, 1.40f, 1.20f, 2.80f, 1.08f, 2.52f, -0.80f, 0.00f, -1.60f,
|
|
0.00f, -1.44f, 0.00f, 0.35f, -1.10f, 0.70f, -2.20f, 0.63f, -1.98f},
|
|
{-0.06f, 0.46f, 0.27f, 0.15f, -0.32f, 0.44f, 0.60f, 1.40f, -1.60f, 0.00f,
|
|
0.63f, -1.98f}};
|
|
if (SingleOpModel::GetForceUseNnapi() && constant_tensors) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
for (size_t i = 0; i < test_shapes.size(); ++i) {
|
|
QuantizedMulOpModel<integer_dtype, integer_dtype> model_fixture(
|
|
{tensor_type, base_shape, -3.f, 3.f},
|
|
{tensor_type, test_shapes[i], -3.f, 3.f}, {tensor_type, {}, -3.f, 3.f},
|
|
ActivationFunctionType_NONE,
|
|
{-0.3f, 2.3f, 0.9f, 0.5f, 0.8f, -1.1f, 1.2f, 2.8f, -1.6f, 0.0f, 0.7f,
|
|
-2.2f},
|
|
{0.2f, 0.3f, -0.4f, 0.5f, 1.0f, 0.9f}, constant_tensors);
|
|
ASSERT_EQ(model_fixture.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(
|
|
model_fixture.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear(test_outputs[i], kQuantizedTolerance)))
|
|
<< "With shape number " << i;
|
|
}
|
|
// Re-run with exchanged inputs.
|
|
for (size_t i = 0; i < test_shapes.size(); ++i) {
|
|
QuantizedMulOpModel<integer_dtype, integer_dtype> model_fixture(
|
|
{tensor_type, test_shapes[i], -3.f, 3.f},
|
|
{tensor_type, base_shape, -3.f, 3.f}, {tensor_type, {}, -3.f, 3.f},
|
|
ActivationFunctionType_NONE, {0.2f, 0.3f, -0.4f, 0.5f, 1.0f, 0.9f},
|
|
{-0.3f, 2.3f, 0.9f, 0.5f, 0.8f, -1.1f, 1.2f, 2.8f, -1.6f, 0.0f, 0.7f,
|
|
-2.2f},
|
|
constant_tensors);
|
|
ASSERT_EQ(model_fixture.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(
|
|
model_fixture.GetDequantizedOutput(),
|
|
ElementsAreArray(ArrayFloatNear(test_outputs[i], kQuantizedTolerance)))
|
|
<< "With shape number " << i;
|
|
}
|
|
}
|
|
|
|
TEST_P(MulOpTest, WithBroadcastUInt8) {
|
|
bool constant_tensors = GetParam();
|
|
WithBroadcast<TensorType_UINT8, uint8_t>(constant_tensors);
|
|
}
|
|
|
|
TEST_P(MulOpTest, WithBroadcastInt8) {
|
|
bool constant_tensors = GetParam();
|
|
WithBroadcast<TensorType_INT8, int8_t>(constant_tensors);
|
|
}
|
|
|
|
TEST_P(MulOpTest, QuantizedWithMixedBroadcastUInt8) {
|
|
bool constant_tensors = GetParam();
|
|
QuantizedWithMixedBroadcast<TensorType_UINT8, uint8_t>(constant_tensors);
|
|
}
|
|
|
|
TEST_P(MulOpTest, QuantizedWithMixedBroadcastInt8) {
|
|
bool constant_tensors = GetParam();
|
|
QuantizedWithMixedBroadcast<TensorType_INT8, int8_t>(constant_tensors);
|
|
}
|
|
|
|
INSTANTIATE_TEST_SUITE_P(ConstantInputs, MulOpTest, testing::Bool());
|
|
|
|
constexpr int kDim1 = 2;
|
|
constexpr int kDim2 = 3;
|
|
constexpr int kDim3 = 4;
|
|
constexpr int kDim4 = 5;
|
|
constexpr int kDim5 = 6;
|
|
constexpr int kDim6 = 7;
|
|
|
|
constexpr int kMaxMulBroadcastDim = 6;
|
|
|
|
template <typename T>
|
|
void TestFloatBroadcast(MulOpModel<T>& m, const std::vector<int>& input1_shape,
|
|
const std::vector<int>& input2_shape) {
|
|
std::array<int, kMaxMulBroadcastDim> input1_dims;
|
|
std::array<int, kMaxMulBroadcastDim> input2_dims;
|
|
std::array<int, kMaxMulBroadcastDim> output_dims;
|
|
std::array<int, kMaxMulBroadcastDim> input1_strides;
|
|
std::array<int, kMaxMulBroadcastDim> input2_strides;
|
|
std::array<int, kMaxMulBroadcastDim> output_strides;
|
|
std::fill(input1_dims.begin(), input1_dims.end(), 1);
|
|
std::fill(input2_dims.begin(), input2_dims.end(), 1);
|
|
std::fill(output_dims.begin(), output_dims.end(), 1);
|
|
std::copy(input1_shape.cbegin(), input1_shape.cend(),
|
|
input1_dims.end() - input1_shape.size());
|
|
std::copy(input2_shape.cbegin(), input2_shape.cend(),
|
|
input2_dims.end() - input2_shape.size());
|
|
|
|
for (size_t i = 0; i < kMaxMulBroadcastDim; i++) {
|
|
if (input1_dims[i] != 1 && input2_dims[i] != 1) {
|
|
ASSERT_EQ(input1_dims[i], input2_dims[i]);
|
|
}
|
|
output_dims[i] = std::max(input1_dims[i], input2_dims[i]);
|
|
}
|
|
// Compute generalized strides.
|
|
size_t input1_stride = 1, input2_stride = 1, output_stride = 1;
|
|
for (size_t i = kMaxMulBroadcastDim; i != 0; i--) {
|
|
input1_strides[i - 1] = input1_dims[i - 1] == 1 ? 0 : input1_stride;
|
|
input2_strides[i - 1] = input2_dims[i - 1] == 1 ? 0 : input2_stride;
|
|
output_strides[i - 1] = output_stride;
|
|
input1_stride *= input1_dims[i - 1];
|
|
input2_stride *= input2_dims[i - 1];
|
|
output_stride *= output_dims[i - 1];
|
|
}
|
|
const int num_input1_elements = std::accumulate(
|
|
input1_dims.begin(), input1_dims.end(), 1, std::multiplies<int>());
|
|
const int num_input2_elements = std::accumulate(
|
|
input2_dims.begin(), input2_dims.end(), 1, std::multiplies<int>());
|
|
const int num_output_elements = std::accumulate(
|
|
output_dims.begin(), output_dims.end(), 1, std::multiplies<int>());
|
|
std::vector<T> input1(num_input1_elements);
|
|
std::vector<T> input2(num_input2_elements);
|
|
std::vector<T> output_ref(num_output_elements);
|
|
|
|
std::random_device random_device;
|
|
auto rng = std::mt19937(random_device());
|
|
std::uniform_real_distribution<float> f32dist(0.01f, 1.0f);
|
|
|
|
std::generate(input1.begin(), input1.end(),
|
|
[&]() { return static_cast<T>(f32dist(rng)); });
|
|
std::generate(input2.begin(), input2.end(),
|
|
[&]() { return static_cast<T>(f32dist(rng)); });
|
|
|
|
// Compute reference results.
|
|
for (size_t i = 0; i < output_dims[0]; i++) {
|
|
for (size_t j = 0; j < output_dims[1]; j++) {
|
|
for (size_t k = 0; k < output_dims[2]; k++) {
|
|
for (size_t l = 0; l < output_dims[3]; l++) {
|
|
for (size_t m = 0; m < output_dims[4]; m++) {
|
|
for (size_t n = 0; n < output_dims[5]; n++) {
|
|
output_ref[i * output_strides[0] + j * output_strides[1] +
|
|
k * output_strides[2] + l * output_strides[3] +
|
|
m * output_strides[4] + n * output_strides[5]] =
|
|
static_cast<T>(
|
|
static_cast<float>(
|
|
input1[i * input1_strides[0] + j * input1_strides[1] +
|
|
k * input1_strides[2] + l * input1_strides[3] +
|
|
m * input1_strides[4] +
|
|
n * input1_strides[5]]) *
|
|
static_cast<float>(
|
|
input2[i * input2_strides[0] + j * input2_strides[1] +
|
|
k * input2_strides[2] + l * input2_strides[3] +
|
|
m * input2_strides[4] +
|
|
n * input2_strides[5]]));
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
m.Resize(input1_shape, input2_shape);
|
|
m.template PopulateTensor<T>(m.input1(), input1);
|
|
m.template PopulateTensor<T>(m.input2(), input2);
|
|
TFLITE_INVOKE_AND_CHECK(T, &m);
|
|
EXPECT_THAT(m.GetOutput(),
|
|
Pointwise(FloatingPointEq(), ToVector<T>(output_ref)));
|
|
}
|
|
|
|
template <typename IntegerType>
|
|
void TestIntegerBroadcast(IntegerMulOpModel<IntegerType>& m,
|
|
const std::vector<int>& input1_shape,
|
|
const std::vector<int>& input2_shape) {
|
|
std::array<int, kMaxMulBroadcastDim> input1_dims;
|
|
std::array<int, kMaxMulBroadcastDim> input2_dims;
|
|
std::array<int, kMaxMulBroadcastDim> output_dims;
|
|
std::array<int, kMaxMulBroadcastDim> input1_strides;
|
|
std::array<int, kMaxMulBroadcastDim> input2_strides;
|
|
std::array<int, kMaxMulBroadcastDim> output_strides;
|
|
std::fill(input1_dims.begin(), input1_dims.end(), 1);
|
|
std::fill(input2_dims.begin(), input2_dims.end(), 1);
|
|
std::fill(output_dims.begin(), output_dims.end(), 1);
|
|
std::copy(input1_shape.cbegin(), input1_shape.cend(),
|
|
input1_dims.end() - input1_shape.size());
|
|
std::copy(input2_shape.cbegin(), input2_shape.cend(),
|
|
input2_dims.end() - input2_shape.size());
|
|
|
|
for (size_t i = 0; i < kMaxMulBroadcastDim; i++) {
|
|
if (input1_dims[i] != 1 && input2_dims[i] != 1) {
|
|
ASSERT_EQ(input1_dims[i], input2_dims[i]);
|
|
}
|
|
output_dims[i] = std::max(input1_dims[i], input2_dims[i]);
|
|
}
|
|
// Compute generalized strides.
|
|
size_t input1_stride = 1, input2_stride = 1, output_stride = 1;
|
|
for (size_t i = kMaxMulBroadcastDim; i != 0; i--) {
|
|
input1_strides[i - 1] = input1_dims[i - 1] == 1 ? 0 : input1_stride;
|
|
input2_strides[i - 1] = input2_dims[i - 1] == 1 ? 0 : input2_stride;
|
|
output_strides[i - 1] = output_stride;
|
|
input1_stride *= input1_dims[i - 1];
|
|
input2_stride *= input2_dims[i - 1];
|
|
output_stride *= output_dims[i - 1];
|
|
}
|
|
const int num_input1_elements = std::accumulate(
|
|
input1_dims.begin(), input1_dims.end(), 1, std::multiplies<int>());
|
|
const int num_input2_elements = std::accumulate(
|
|
input2_dims.begin(), input2_dims.end(), 1, std::multiplies<int>());
|
|
const int num_output_elements = std::accumulate(
|
|
output_dims.begin(), output_dims.end(), 1, std::multiplies<int>());
|
|
std::vector<IntegerType> input1(num_input1_elements);
|
|
std::vector<IntegerType> input2(num_input2_elements);
|
|
std::vector<IntegerType> output_ref(num_output_elements);
|
|
|
|
std::random_device random_device;
|
|
auto rng = std::mt19937(random_device());
|
|
std::uniform_int_distribution<IntegerType> dist(0, 256);
|
|
|
|
std::generate(input1.begin(), input1.end(), [&]() { return dist(rng); });
|
|
std::generate(input2.begin(), input2.end(), [&]() { return dist(rng); });
|
|
|
|
// Compute reference results.
|
|
for (size_t i = 0; i < output_dims[0]; i++) {
|
|
for (size_t j = 0; j < output_dims[1]; j++) {
|
|
for (size_t k = 0; k < output_dims[2]; k++) {
|
|
for (size_t l = 0; l < output_dims[3]; l++) {
|
|
for (size_t m = 0; m < output_dims[4]; m++) {
|
|
for (size_t n = 0; n < output_dims[5]; n++) {
|
|
output_ref[i * output_strides[0] + j * output_strides[1] +
|
|
k * output_strides[2] + l * output_strides[3] +
|
|
m * output_strides[4] + n * output_strides[5]] =
|
|
input1[i * input1_strides[0] + j * input1_strides[1] +
|
|
k * input1_strides[2] + l * input1_strides[3] +
|
|
m * input1_strides[4] + n * input1_strides[5]] *
|
|
input2[i * input2_strides[0] + j * input2_strides[1] +
|
|
k * input2_strides[2] + l * input2_strides[3] +
|
|
m * input2_strides[4] + n * input2_strides[5]];
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
m.Resize(input1_shape, input2_shape);
|
|
m.template PopulateTensor<IntegerType>(m.input1(), input1);
|
|
m.template PopulateTensor<IntegerType>(m.input2(), input2);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
EXPECT_THAT(m.GetOutput(), Pointwise(FloatingPointEq(), output_ref));
|
|
}
|
|
|
|
// To improve automatic test sharding (via shard_count in the BUILD file),
|
|
// we need to ensure that each individual test case runs in a reasonable time,
|
|
// otherwise we end up being limited by the performance of the longest shard.
|
|
// Since TestFloat32MultiDimBroadcast has 2^12 iterations, it takes a
|
|
// long time (over 30 seconds) to execute all iterations -- too long for a
|
|
// single shard. So we split it into a few \"subshards\" and have a separate
|
|
// TYPED_TEST macro invocation for each subshard.
|
|
|
|
template <typename T>
|
|
void RunFloatMultiDimBroadcastTest(int d1, int d2) {
|
|
const int dims_constants[] = {kDim1, kDim2, kDim3, kDim4, kDim5, kDim6};
|
|
std::vector<int> initial_shape1(d1, 1);
|
|
std::vector<int> initial_shape2(d2, 1);
|
|
MulOpModel<T> m({GetTensorType<T>(), initial_shape1},
|
|
{GetTensorType<T>(), initial_shape2},
|
|
{GetTensorType<T>(), {}}, ActivationFunctionType_NONE, {}, {},
|
|
/*constant_tensors=*/false);
|
|
|
|
for (uint32_t bm1 = 0; bm1 < (static_cast<uint32_t>(1) << d1); bm1++) {
|
|
for (uint32_t bm2 = 0; bm2 < (static_cast<uint32_t>(1) << d2); bm2++) {
|
|
std::vector<int> input1_shape(d1);
|
|
std::vector<int> input2_shape(d2);
|
|
for (int i = 0; i < d1; ++i) {
|
|
bool broadcast = bm1 & (1 << i);
|
|
input1_shape[i] = broadcast ? 1 : dims_constants[6 - d1 + i];
|
|
}
|
|
for (int i = 0; i < d2; ++i) {
|
|
bool broadcast = bm2 & (1 << i);
|
|
input2_shape[i] = broadcast ? 1 : dims_constants[6 - d2 + i];
|
|
}
|
|
TestFloatBroadcast<T>(m, input1_shape, input2_shape);
|
|
if (testing::Test::IsSkipped()) {
|
|
return;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
#define INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, d2) \
|
|
TYPED_TEST(FloatMulTest, MultiDimBroadcast_##d1##_##d2) { \
|
|
RunFloatMultiDimBroadcastTest<TypeParam>(d1, d2); \
|
|
}
|
|
|
|
#define INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(d1) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, 1) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, 2) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, 3) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, 4) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, 5) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST(d1, 6)
|
|
|
|
#define INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TESTS() \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(1) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(2) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(3) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(4) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(5) \
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TEST_D2(6)
|
|
|
|
INSTANTIATE_FLOAT_MUL_MULTI_DIM_BROADCAST_TESTS()
|
|
|
|
template <typename T>
|
|
class IntegerMulOpTest : public ::testing::Test {};
|
|
|
|
using Int16OrInt32Or64Types = ::testing::Types<int16_t, int32_t, int64_t>;
|
|
TYPED_TEST_SUITE(IntegerMulOpTest, Int16OrInt32Or64Types);
|
|
|
|
// To improve automatic test sharding (via shard_count in the BUILD file),
|
|
// we need to ensure that each individual test case runs in a reasonable time,
|
|
// otherwise we end up being limited by the performance of the longest shard.
|
|
// Since TestIntegerMultiDimBroadcast has 2^12 iterations, it takes a
|
|
// long time (over 30 seconds) to execute all iterations -- too long for a
|
|
// single shard. So we split it into a few "subshards" and have a separate
|
|
// TYPED_TEST macro invocation for each subshard.
|
|
|
|
template <typename TypeParam>
|
|
void RunIntegerMultiDimBroadcastTest(int d1, int d2) {
|
|
const int dims_constants[] = {kDim1, kDim2, kDim3, kDim4, kDim5, kDim6};
|
|
std::vector<int> initial_shape1(d1, 1);
|
|
std::vector<int> initial_shape2(d2, 1);
|
|
IntegerMulOpModel<TypeParam> m({GetTensorType<TypeParam>(), initial_shape1},
|
|
{GetTensorType<TypeParam>(), initial_shape2},
|
|
{GetTensorType<TypeParam>(), {}},
|
|
ActivationFunctionType_NONE, {}, {},
|
|
/*constant_tensors=*/false);
|
|
|
|
for (uint32_t bm1 = 0; bm1 < (static_cast<uint32_t>(1) << d1); bm1++) {
|
|
for (uint32_t bm2 = 0; bm2 < (static_cast<uint32_t>(1) << d2); bm2++) {
|
|
std::vector<int> input1_shape(d1);
|
|
std::vector<int> input2_shape(d2);
|
|
for (int i = 0; i < d1; ++i) {
|
|
bool broadcast = bm1 & (1 << i);
|
|
input1_shape[i] = broadcast ? 1 : dims_constants[6 - d1 + i];
|
|
}
|
|
for (int i = 0; i < d2; ++i) {
|
|
bool broadcast = bm2 & (1 << i);
|
|
input2_shape[i] = broadcast ? 1 : dims_constants[6 - d2 + i];
|
|
}
|
|
TestIntegerBroadcast<TypeParam>(m, input1_shape, input2_shape);
|
|
if (testing::Test::IsSkipped()) {
|
|
return;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
#define INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, d2) \
|
|
TYPED_TEST(IntegerMulOpTest, MultiDimBroadcast_##d1##_##d2) { \
|
|
RunIntegerMultiDimBroadcastTest<TypeParam>(d1, d2); \
|
|
}
|
|
|
|
#define INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(d1) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, 1) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, 2) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, 3) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, 4) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, 5) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST(d1, 6)
|
|
|
|
#define INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TESTS() \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(1) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(2) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(3) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(4) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(5) \
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TEST_D2(6)
|
|
|
|
INSTANTIATE_INTEGER_MUL_MULTI_DIM_BROADCAST_TESTS()
|
|
|
|
template <typename QuantizedType>
|
|
void TestQuantizedBroadcast(
|
|
QuantizedMulOpModel<QuantizedType, QuantizedType>& m,
|
|
const std::vector<int>& input1_shape,
|
|
const std::vector<int>& input2_shape) {
|
|
std::array<int, kMaxMulBroadcastDim> input1_dims;
|
|
std::array<int, kMaxMulBroadcastDim> input2_dims;
|
|
std::array<int, kMaxMulBroadcastDim> output_dims;
|
|
std::array<int, kMaxMulBroadcastDim> input1_strides;
|
|
std::array<int, kMaxMulBroadcastDim> input2_strides;
|
|
std::array<int, kMaxMulBroadcastDim> output_strides;
|
|
std::fill(input1_dims.begin(), input1_dims.end(), 1);
|
|
std::fill(input2_dims.begin(), input2_dims.end(), 1);
|
|
std::fill(output_dims.begin(), output_dims.end(), 1);
|
|
std::copy(input1_shape.cbegin(), input1_shape.cend(),
|
|
input1_dims.end() - input1_shape.size());
|
|
std::copy(input2_shape.cbegin(), input2_shape.cend(),
|
|
input2_dims.end() - input2_shape.size());
|
|
|
|
for (size_t i = 0; i < kMaxMulBroadcastDim; i++) {
|
|
if (input1_dims[i] != 1 && input2_dims[i] != 1) {
|
|
ASSERT_EQ(input1_dims[i], input2_dims[i]);
|
|
}
|
|
output_dims[i] = std::max(input1_dims[i], input2_dims[i]);
|
|
}
|
|
// Compute generalized strides.
|
|
size_t input1_stride = 1, input2_stride = 1, output_stride = 1;
|
|
for (size_t i = kMaxMulBroadcastDim; i != 0; i--) {
|
|
input1_strides[i - 1] = input1_dims[i - 1] == 1 ? 0 : input1_stride;
|
|
input2_strides[i - 1] = input2_dims[i - 1] == 1 ? 0 : input2_stride;
|
|
output_strides[i - 1] = output_stride;
|
|
input1_stride *= input1_dims[i - 1];
|
|
input2_stride *= input2_dims[i - 1];
|
|
output_stride *= output_dims[i - 1];
|
|
}
|
|
const int num_input1_elements = std::accumulate(
|
|
input1_dims.begin(), input1_dims.end(), 1, std::multiplies<int>());
|
|
const int num_input2_elements = std::accumulate(
|
|
input2_dims.begin(), input2_dims.end(), 1, std::multiplies<int>());
|
|
const int num_output_elements = std::accumulate(
|
|
output_dims.begin(), output_dims.end(), 1, std::multiplies<int>());
|
|
std::vector<float> input1(num_input1_elements);
|
|
std::vector<float> input2(num_input2_elements);
|
|
std::vector<float> output_ref(num_output_elements);
|
|
|
|
std::random_device random_device;
|
|
auto rng = std::mt19937(random_device());
|
|
|
|
std::uniform_real_distribution<float> dist(-0.5f, 0.5f);
|
|
|
|
std::generate(input1.begin(), input1.end(), [&]() { return dist(rng); });
|
|
std::generate(input2.begin(), input2.end(), [&]() { return dist(rng); });
|
|
|
|
m.Resize(input1_shape, input2_shape);
|
|
m.template QuantizeAndPopulate<QuantizedType>(m.input1(), input1);
|
|
m.template QuantizeAndPopulate<QuantizedType>(m.input2(), input2);
|
|
// Compute reference results.
|
|
for (size_t i = 0; i < output_dims[0]; i++) {
|
|
for (size_t j = 0; j < output_dims[1]; j++) {
|
|
for (size_t k = 0; k < output_dims[2]; k++) {
|
|
for (size_t l = 0; l < output_dims[3]; l++) {
|
|
for (size_t m = 0; m < output_dims[4]; m++) {
|
|
for (size_t n = 0; n < output_dims[5]; n++) {
|
|
float x = input1[i * input1_strides[0] + j * input1_strides[1] +
|
|
k * input1_strides[2] + l * input1_strides[3] +
|
|
m * input1_strides[4] + n * input1_strides[5]];
|
|
float y = input2[i * input2_strides[0] + j * input2_strides[1] +
|
|
k * input2_strides[2] + l * input2_strides[3] +
|
|
m * input2_strides[4] + n * input2_strides[5]];
|
|
output_ref[i * output_strides[0] + j * output_strides[1] +
|
|
k * output_strides[2] + l * output_strides[3] +
|
|
m * output_strides[4] + n * output_strides[5]] = x * y;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
for (float& output_value : output_ref) {
|
|
output_value = std::max<float>(output_value, -1.0f);
|
|
output_value = std::min<float>(output_value, 1.0f);
|
|
}
|
|
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
std::vector<float> output = m.GetDequantizedOutput();
|
|
for (size_t i = 0; i < output_dims[0]; i++) {
|
|
for (size_t j = 0; j < output_dims[1]; j++) {
|
|
for (size_t k = 0; k < output_dims[2]; k++) {
|
|
for (size_t l = 0; l < output_dims[3]; l++) {
|
|
for (size_t m = 0; m < output_dims[4]; m++) {
|
|
for (size_t n = 0; n < output_dims[5]; n++) {
|
|
const size_t index =
|
|
i * output_strides[0] + j * output_strides[1] +
|
|
k * output_strides[2] + l * output_strides[3] +
|
|
m * output_strides[4] + n * output_strides[5];
|
|
EXPECT_NEAR(output[index], output_ref[index], 0.6f)
|
|
<< "(i, j, k, l, m, n) = (" << i << ", " << j << ", " << k
|
|
<< ", " << l << ", " << m << ", " << n << ")";
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
// To improve automatic test sharding (via shard_count in the BUILD file),
|
|
// we need to ensure that each individual test case runs in a reasonable time,
|
|
// otherwise we end up being limited by the performance of the longest shard.
|
|
// Since TestQuantizedMultiDimBroadcast has 2^12 iterations, it takes a
|
|
// long time (over 30 seconds) to execute all iterations -- too long for a
|
|
// single shard. So we split it into a few "subshards" and have a separate
|
|
// TEST macro invocation for each subshard.
|
|
|
|
template <typename T>
|
|
void RunQuantizedMultiDimBroadcastTest(int d1, int d2) {
|
|
const int dims_constants[] = {kDim1, kDim2, kDim3, kDim4, kDim5, kDim6};
|
|
std::vector<int> initial_shape1(d1, 1);
|
|
std::vector<int> initial_shape2(d2, 1);
|
|
QuantizedMulOpModel<T, T> m({GetTensorType<T>(), initial_shape1, -0.5f, 0.5f},
|
|
{GetTensorType<T>(), initial_shape2, -0.5f, 0.5f},
|
|
{GetTensorType<T>(), {}, -1.f, 1.f},
|
|
ActivationFunctionType_NONE, {}, {},
|
|
/*constant_tensors=*/false);
|
|
|
|
for (uint32_t bm1 = 0; bm1 < (static_cast<uint32_t>(1) << d1); bm1++) {
|
|
for (uint32_t bm2 = 0; bm2 < (static_cast<uint32_t>(1) << d2); bm2++) {
|
|
std::vector<int> input1_shape(d1);
|
|
std::vector<int> input2_shape(d2);
|
|
for (int i = 0; i < d1; ++i) {
|
|
bool broadcast = bm1 & (1 << i);
|
|
input1_shape[i] = broadcast ? 1 : dims_constants[6 - d1 + i];
|
|
}
|
|
for (int i = 0; i < d2; ++i) {
|
|
bool broadcast = bm2 & (1 << i);
|
|
input2_shape[i] = broadcast ? 1 : dims_constants[6 - d2 + i];
|
|
}
|
|
TestQuantizedBroadcast<T>(m, input1_shape, input2_shape);
|
|
if (testing::Test::IsSkipped()) {
|
|
return;
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
#define INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, \
|
|
d2) \
|
|
TEST(QuantizedMulOpModel, \
|
|
TypeName##QuantizedMultiDimBroadcast_##d1##_##d2) { \
|
|
RunQuantizedMultiDimBroadcastTest<T>(d1, d2); \
|
|
}
|
|
|
|
#define INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, d1) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, 1) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, 2) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, 3) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, 4) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, 5) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST(T, TypeName, d1, 6)
|
|
|
|
#define INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TESTS(T, TypeName) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, 1) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, 2) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, 3) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, 4) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, 5) \
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TEST_D2(T, TypeName, 6)
|
|
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TESTS(int8_t, Int8)
|
|
INSTANTIATE_QUANTIZED_MUL_MULTI_DIM_BROADCAST_TESTS(uint8_t, Uint8)
|
|
|
|
} // namespace
|
|
} // namespace tflite
|