wehub-resource-sync
361 lines
13 KiB
C++
361 lines
13 KiB
C++
/* Copyright 2017 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 <stdint.h>
|
|
|
|
#include <initializer_list>
|
|
#include <string>
|
|
#include <type_traits>
|
|
#include <vector>
|
|
|
|
#include <gmock/gmock.h>
|
|
#include <gtest/gtest.h>
|
|
#include "tensorflow/lite/kernels/test_util.h"
|
|
#include "tensorflow/lite/schema/schema_generated.h"
|
|
#include "tensorflow/lite/string_type.h"
|
|
|
|
namespace tflite {
|
|
namespace {
|
|
|
|
using ::testing::ElementsAreArray;
|
|
class TileOpBaseModel : public SingleOpModel {
|
|
public:
|
|
template <typename T>
|
|
void SetInput(std::initializer_list<T> data) {
|
|
PopulateTensor<T>(input_, data);
|
|
}
|
|
|
|
template <typename T>
|
|
void SetMultipliers(std::initializer_list<T> data) {
|
|
PopulateTensor<T>(multipliers_, data);
|
|
}
|
|
|
|
template <typename T>
|
|
std::vector<T> GetOutput() {
|
|
return ExtractVector<T>(output_);
|
|
}
|
|
|
|
std::vector<int> GetOutputShape() { return GetTensorShape(output_); }
|
|
|
|
protected:
|
|
int input_;
|
|
int multipliers_;
|
|
int output_;
|
|
};
|
|
|
|
template <typename InputType, typename MultipliersType = int32_t>
|
|
class TileOpConstModel : public TileOpBaseModel {
|
|
public:
|
|
TileOpConstModel(std::initializer_list<int> input_shape,
|
|
std::initializer_list<InputType> input_data,
|
|
TensorType input_type, TensorType multiply_type,
|
|
std::initializer_list<MultipliersType> multipliers_data,
|
|
bool allocate_and_delegate = true) {
|
|
SetupInput(input_shape, input_data, input_type,
|
|
std::is_same<std::string, InputType>());
|
|
multipliers_ = AddConstInput(multiply_type, multipliers_data,
|
|
{static_cast<int>(multipliers_data.size())});
|
|
output_ = AddOutput(input_type);
|
|
SetBuiltinOp(BuiltinOperator_TILE, BuiltinOptions_TileOptions, 0);
|
|
BuildInterpreter({input_shape, {static_cast<int>(input_shape.size())}},
|
|
/*num_threads=*/-1,
|
|
/*allow_fp32_relax_to_fp16=*/false,
|
|
/*apply_delegate=*/true, allocate_and_delegate);
|
|
if (allocate_and_delegate) {
|
|
PopulateInput(input_data, std::is_same<std::string, InputType>());
|
|
}
|
|
}
|
|
|
|
private:
|
|
template <typename T>
|
|
void SetupInput(std::initializer_list<int> input_shape,
|
|
std::initializer_list<T> input_data, TensorType input_type,
|
|
std::false_type) {
|
|
input_ = AddConstInput(input_type, input_data, input_shape);
|
|
}
|
|
template <typename T>
|
|
void SetupInput(std::initializer_list<int> input_shape,
|
|
std::initializer_list<T> input_data, TensorType input_type,
|
|
std::true_type) {
|
|
input_ = AddInput(input_type);
|
|
}
|
|
template <typename T>
|
|
void PopulateInput(std::initializer_list<T> input_data, std::false_type) {}
|
|
template <typename T>
|
|
void PopulateInput(std::initializer_list<T> input_data, std::true_type) {
|
|
SetInput(input_data);
|
|
}
|
|
};
|
|
|
|
class TileOpDynamicModel : public TileOpBaseModel {
|
|
public:
|
|
TileOpDynamicModel(std::initializer_list<int> input_shape,
|
|
TensorType input_type, TensorType multiply_type) {
|
|
input_ = AddInput(input_type);
|
|
multipliers_ = AddInput(multiply_type);
|
|
output_ = AddOutput(input_type);
|
|
SetBuiltinOp(BuiltinOperator_TILE, BuiltinOptions_TileOptions, 0);
|
|
BuildInterpreter({input_shape, {static_cast<int>(input_shape.size())}});
|
|
}
|
|
};
|
|
|
|
enum class TestType {
|
|
kConst = 0,
|
|
kDynamic = 1,
|
|
};
|
|
|
|
template <typename InputType, typename MultipliersType = int32_t>
|
|
void Check(std::initializer_list<int> input_shape,
|
|
std::initializer_list<InputType> input_data,
|
|
std::initializer_list<MultipliersType> multipliers_data,
|
|
std::initializer_list<int> exp_output_shape,
|
|
std::initializer_list<InputType> exp_output_data,
|
|
TensorType input_type, TensorType multiply_type,
|
|
TestType test_type) {
|
|
switch (test_type) {
|
|
case TestType::kConst: {
|
|
if (SingleOpModel::GetForceUseNnapi() &&
|
|
!std::is_same<InputType, std::string>::value) {
|
|
// NNAPI does not support graphs with all constant inputs.
|
|
return;
|
|
}
|
|
TileOpConstModel<InputType, MultipliersType> m(
|
|
input_shape, input_data, input_type, multiply_type, multipliers_data);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
|
|
EXPECT_THAT(m.GetOutputShape(), ElementsAreArray(exp_output_shape));
|
|
EXPECT_THAT(m.template GetOutput<InputType>(),
|
|
ElementsAreArray(exp_output_data));
|
|
return;
|
|
}
|
|
case TestType::kDynamic: {
|
|
TileOpDynamicModel m(input_shape, input_type, multiply_type);
|
|
m.SetInput(input_data);
|
|
m.SetMultipliers(multipliers_data);
|
|
ASSERT_EQ(m.Invoke(), kTfLiteOk);
|
|
|
|
EXPECT_THAT(m.GetOutputShape(), ElementsAreArray(exp_output_shape));
|
|
EXPECT_THAT(m.template GetOutput<InputType>(),
|
|
ElementsAreArray(exp_output_data));
|
|
return;
|
|
}
|
|
}
|
|
}
|
|
|
|
class TileTest : public ::testing::TestWithParam<TestType> {};
|
|
|
|
TEST_P(TileTest, Float32Vector) {
|
|
Check<float>(/*input_shape=*/{3},
|
|
/*input_data=*/{1.0, 2.0, 3.0},
|
|
/*multipliers_data=*/{2}, /*exp_output_shape=*/{6},
|
|
/*exp_output_data=*/{1.0, 2.0, 3.0, 1.0, 2.0, 3.0},
|
|
/*input_type=*/TensorType_FLOAT32,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Float32Matrix) {
|
|
Check<float>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{11.f, 12.f, 13.f, 21.f, 22.f, 23.f},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/
|
|
{11.f, 12.f, 13.f, 21.f, 22.f, 23.f, 11.f, 12.f, 13.f, 21.f, 22.f, 23.f},
|
|
/*input_type=*/TensorType_FLOAT32,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Float32HighDimension) {
|
|
Check<float>(
|
|
/*input_shape=*/{1, 2, 3},
|
|
/*input_data=*/{11.f, 12.f, 13.f, 21.f, 22.f, 23.f},
|
|
/*multipliers_data=*/{2, 3, 1}, /*exp_output_shape=*/{2, 6, 3},
|
|
/*exp_output_data=*/{11.f, 12.f, 13.f, 21.f, 22.f, 23.f, 11.f, 12.f,
|
|
13.f, 21.f, 22.f, 23.f, 11.f, 12.f, 13.f, 21.f,
|
|
22.f, 23.f, 11.f, 12.f, 13.f, 21.f, 22.f, 23.f,
|
|
11.f, 12.f, 13.f, 21.f, 22.f, 23.f, 11.f, 12.f,
|
|
13.f, 21.f, 22.f, 23.f},
|
|
/*input_type=*/TensorType_FLOAT32,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Uint8Matrix) {
|
|
Check<uint8_t>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{11, 12, 13, 21, 22, 23},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/{11, 12, 13, 21, 22, 23, 11, 12, 13, 21, 22, 23},
|
|
/*input_type=*/TensorType_UINT8,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Int32Matrix) {
|
|
Check<int32_t>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{11, 12, 13, 21, 22, 23},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/{11, 12, 13, 21, 22, 23, 11, 12, 13, 21, 22, 23},
|
|
/*input_type=*/TensorType_INT32,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, BooleanMatrix) {
|
|
Check<bool>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{true, false, false, true, true, false},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/
|
|
{true, false, false, true, true, false, true, false, false, true, true,
|
|
false},
|
|
/*input_type=*/TensorType_BOOL,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Int64Matrix) {
|
|
Check<int64_t>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{11, 12, 13, 21, 22, 23},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/{11, 12, 13, 21, 22, 23, 11, 12, 13, 21, 22, 23},
|
|
/*input_type=*/TensorType_INT64,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Int64Matrix64Multipliers) {
|
|
Check<int64_t, int64_t>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{11, 12, 13, 21, 22, 23},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/{11, 12, 13, 21, 22, 23, 11, 12, 13, 21, 22, 23},
|
|
/*input_type=*/TensorType_INT64,
|
|
/*multiply_type=*/TensorType_INT64, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Int8Matrix) {
|
|
if (SingleOpModel::GetForceUseNnapi()) {
|
|
return;
|
|
}
|
|
Check<int8_t>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{11, 12, 13, 21, 22, 23},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/{11, 12, 13, 21, 22, 23, 11, 12, 13, 21, 22, 23},
|
|
/*input_type=*/TensorType_INT8,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, StringMatrix) {
|
|
Check<std::string>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{"AA", "AB", "AC", "BA", "BB", "BC"},
|
|
/*multipliers_data=*/{1, 2}, /*exp_output_shape=*/{2, 6},
|
|
/*exp_output_data=*/
|
|
{"AA", "AB", "AC", "AA", "AB", "AC", "BA", "BB", "BC", "BA", "BB", "BC"},
|
|
/*input_type=*/TensorType_STRING,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, StringMatrix64Multipliers) {
|
|
Check<std::string, int64_t>(
|
|
/*input_shape=*/{2, 3},
|
|
/*input_data=*/{"AA", "AB", "AC", "BA", "BB", "BC"},
|
|
/*multipliers_data=*/{2, 1}, /*exp_output_shape=*/{4, 3},
|
|
/*exp_output_data=*/
|
|
{"AA", "AB", "AC", "BA", "BB", "BC", "AA", "AB", "AC", "BA", "BB", "BC"},
|
|
/*input_type=*/TensorType_STRING,
|
|
/*multiply_type=*/TensorType_INT64, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, StringMatrix2) {
|
|
Check<std::string>(
|
|
/*input_shape=*/{3, 2, 1},
|
|
/*input_data=*/{"AA", "AB", "AC", "BA", "BB", "BC"},
|
|
/*multipliers_data=*/{2, 2, 2}, /*exp_output_shape=*/{6, 4, 2},
|
|
/*exp_output_data=*/
|
|
{"AA", "AA", "AB", "AB", "AA", "AA", "AB", "AB", "AC", "AC", "BA", "BA",
|
|
"AC", "AC", "BA", "BA", "BB", "BB", "BC", "BC", "BB", "BB", "BC", "BC",
|
|
"AA", "AA", "AB", "AB", "AA", "AA", "AB", "AB", "AC", "AC", "BA", "BA",
|
|
"AC", "AC", "BA", "BA", "BB", "BB", "BC", "BC", "BB", "BB", "BC", "BC"},
|
|
/*input_type=*/TensorType_STRING,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, StringMatrixEmptyInputElements) {
|
|
Check<std::string>(
|
|
/*input_shape=*/{0, 1, 1},
|
|
/*input_data=*/{},
|
|
/*multipliers_data=*/{2, 2, 2}, /*exp_output_shape=*/{0, 2, 2},
|
|
/*exp_output_data=*/
|
|
{},
|
|
/*input_type=*/TensorType_STRING,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST_P(TileTest, Int32EmptyInput) {
|
|
Check<int32_t>(/*input_shape=*/{2, 1, 3},
|
|
/*input_data=*/{11, 12, 13, 21, 22, 23},
|
|
/*multipliers_data=*/{2, 0, 2},
|
|
/*exp_output_shape=*/{4, 0, 6},
|
|
/*exp_output_data=*/{},
|
|
/*input_type=*/TensorType_INT32,
|
|
/*multiply_type=*/TensorType_INT32, GetParam());
|
|
}
|
|
|
|
TEST(TileDynamicTest, NegativeMultipliersString) {
|
|
TileOpDynamicModel m({2, 1, 1}, TensorType_STRING, TensorType_INT32);
|
|
m.SetInput<std::string>({"A", "B"});
|
|
m.SetMultipliers({2, -100, -10});
|
|
EXPECT_NE(m.Invoke(), kTfLiteOk);
|
|
}
|
|
|
|
template <typename MultipliersType>
|
|
void CheckInvalidMultipliers(
|
|
TensorType multiply_type, TestType test_type,
|
|
std::initializer_list<MultipliersType> multipliers_data) {
|
|
switch (test_type) {
|
|
case TestType::kConst: {
|
|
TileOpConstModel<int32_t, MultipliersType> m(
|
|
{2, 3}, {11, 12, 13, 21, 22, 23}, TensorType_INT32, multiply_type,
|
|
multipliers_data, /*allocate_and_delegate=*/false);
|
|
EXPECT_NE(m.AllocateTensors(), kTfLiteOk);
|
|
return;
|
|
}
|
|
case TestType::kDynamic: {
|
|
TileOpDynamicModel m({2, 3}, TensorType_INT32, multiply_type);
|
|
m.SetInput({11, 12, 13, 21, 22, 23});
|
|
m.SetMultipliers(multipliers_data);
|
|
EXPECT_NE(m.Invoke(), kTfLiteOk);
|
|
return;
|
|
}
|
|
}
|
|
}
|
|
|
|
TEST_P(TileTest, MultiplierOverflowInt32) {
|
|
CheckInvalidMultipliers<int32_t>(TensorType_INT32, GetParam(),
|
|
{1073741824, 1});
|
|
}
|
|
|
|
TEST_P(TileTest, MultiplierNegativeInt32) {
|
|
CheckInvalidMultipliers<int32_t>(TensorType_INT32, GetParam(), {-1, 1});
|
|
}
|
|
|
|
TEST_P(TileTest, MultiplierNegativeInt64) {
|
|
CheckInvalidMultipliers<int64_t>(TensorType_INT64, GetParam(), {-1LL, 1LL});
|
|
}
|
|
|
|
INSTANTIATE_TEST_SUITE_P(TileTest, TileTest,
|
|
::testing::Values(TestType::kConst,
|
|
TestType::kDynamic));
|
|
} // namespace
|
|
} // namespace tflite
|