170 lines
7.0 KiB
C++
170 lines
7.0 KiB
C++
/* Copyright 2019 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 "tensorflow/lite/c/builtin_op_data.h"
|
|
#include "tensorflow/lite/c/common.h"
|
|
#include "tensorflow/lite/micro/all_ops_resolver.h"
|
|
#include "tensorflow/lite/micro/kernels/kernel_runner.h"
|
|
#include "tensorflow/lite/micro/test_helpers.h"
|
|
#include "tensorflow/lite/micro/testing/micro_test.h"
|
|
|
|
namespace tflite {
|
|
namespace testing {
|
|
namespace {
|
|
|
|
// The Logistic kernel assumes an output in the range [0, 1.0], leading to these
|
|
// quantization parameters.
|
|
const float quantized_output_scale = 1.0 / 255.0;
|
|
const int quantized_output_zero_point_int8 = -128;
|
|
|
|
const int flat_size_basic = 10;
|
|
const int shape_basic[] = {2, 2, 5};
|
|
const float input_data_basic[] = {1, 2, 3, 4, 5, -1, -2, -3, -4, -5};
|
|
const float golden_basic[] = {0.73105858, 0.88079708, 0.95257413, 0.98201379,
|
|
0.99330715, 0.26894142, 0.11920292, 0.04742587,
|
|
0.01798621, 0.00669285};
|
|
|
|
const int flat_size_wide_range = 10;
|
|
const int shape_wide_range[] = {2, 1, 5};
|
|
const float input_data_wide_range[]{
|
|
1.0, 2.0, 3.0, 4.0, 93.0, -1.0, -2.0, -3.0, -4.0, -93.0,
|
|
};
|
|
const float golden_wide_range[] = {
|
|
0.73105858, 0.88079708, 0.95257413, 0.98201379, 1.0,
|
|
0.26894142, 0.11920292, 0.04742587, 0.01798621, 0.0,
|
|
};
|
|
|
|
template <typename T>
|
|
void ValidateLogisticGoldens(TfLiteTensor* tensors, const int tensor_count,
|
|
T* output_data, const T* golden,
|
|
int output_dims_count, float tolerance) {
|
|
int inputs_array_data[] = {1, 0};
|
|
TfLiteIntArray* inputs_array = IntArrayFromInts(inputs_array_data);
|
|
int outputs_array_data[] = {1, 1};
|
|
TfLiteIntArray* outputs_array = IntArrayFromInts(outputs_array_data);
|
|
|
|
const TfLiteRegistration registration =
|
|
tflite::ops::micro::Register_LOGISTIC();
|
|
micro::KernelRunner runner(registration, tensors, tensor_count, inputs_array,
|
|
outputs_array, nullptr, micro_test::reporter);
|
|
|
|
TF_LITE_MICRO_EXPECT_EQ(kTfLiteOk, runner.InitAndPrepare());
|
|
TF_LITE_MICRO_EXPECT_EQ(kTfLiteOk, runner.Invoke());
|
|
|
|
for (int i = 0; i < output_dims_count; ++i) {
|
|
TF_LITE_MICRO_EXPECT_NEAR(golden[i], output_data[i], tolerance);
|
|
}
|
|
}
|
|
|
|
void TestLogisticFloat(const int* input_dims_data, const float* input_data,
|
|
const float* golden, const int* output_dims_data,
|
|
float* output_data) {
|
|
TfLiteIntArray* input_dims = IntArrayFromInts(input_dims_data);
|
|
TfLiteIntArray* output_dims = IntArrayFromInts(output_dims_data);
|
|
const int output_elements_count = ElementCount(*output_dims);
|
|
|
|
constexpr int inputs_size = 1;
|
|
constexpr int outputs_size = 1;
|
|
constexpr int tensors_size = inputs_size + outputs_size;
|
|
TfLiteTensor tensors[tensors_size] = {
|
|
CreateTensor(input_data, input_dims),
|
|
CreateTensor(output_data, output_dims),
|
|
};
|
|
|
|
ValidateLogisticGoldens(tensors, tensors_size, output_data, golden,
|
|
output_elements_count, 1e-5);
|
|
}
|
|
|
|
template <typename T>
|
|
void TestLogisticQuantized(const int* input_dims_data, const float* input_data,
|
|
T* input_quantized, const float input_scale,
|
|
const int input_zero_point, const float* golden,
|
|
T* golden_quantized, const int* output_dims_data,
|
|
const float output_scale,
|
|
const int output_zero_point, int8_t* output_data) {
|
|
TfLiteIntArray* input_dims = IntArrayFromInts(input_dims_data);
|
|
TfLiteIntArray* output_dims = IntArrayFromInts(output_dims_data);
|
|
const int output_elements_count = ElementCount(*output_dims);
|
|
|
|
constexpr int inputs_size = 1;
|
|
constexpr int outputs_size = 1;
|
|
constexpr int tensors_size = inputs_size + outputs_size;
|
|
TfLiteTensor tensors[tensors_size] = {
|
|
CreateQuantizedTensor(input_data, input_quantized, input_dims,
|
|
input_scale, input_zero_point),
|
|
CreateQuantizedTensor(output_data, output_dims, output_scale,
|
|
output_zero_point),
|
|
};
|
|
|
|
tflite::Quantize(golden, golden_quantized, output_elements_count,
|
|
output_scale, output_zero_point);
|
|
ValidateLogisticGoldens(tensors, tensors_size, output_data, golden_quantized,
|
|
output_elements_count, 1.0);
|
|
}
|
|
|
|
} // namespace
|
|
} // namespace testing
|
|
} // namespace tflite
|
|
|
|
TF_LITE_MICRO_TESTS_BEGIN
|
|
|
|
TF_LITE_MICRO_TEST(LogisticFloatBasicShouldMatchGolden) {
|
|
float output_data[tflite::testing::flat_size_basic];
|
|
tflite::testing::TestLogisticFloat(
|
|
tflite::testing::shape_basic, tflite::testing::input_data_basic,
|
|
tflite::testing::golden_basic, tflite::testing::shape_basic, output_data);
|
|
}
|
|
|
|
TF_LITE_MICRO_TEST(LogisticQuantizedInt8BasicShouldMatchGolden) {
|
|
const float input_scale = 0.1;
|
|
const int input_zero_point = 0;
|
|
int8_t input_quantized[tflite::testing::flat_size_basic];
|
|
int8_t golden_quantized[tflite::testing::flat_size_basic];
|
|
int8_t output_data[tflite::testing::flat_size_basic];
|
|
|
|
tflite::testing::TestLogisticQuantized(
|
|
tflite::testing::shape_basic, tflite::testing::input_data_basic,
|
|
input_quantized, input_scale, input_zero_point,
|
|
tflite::testing::golden_basic, golden_quantized,
|
|
tflite::testing::shape_basic, tflite::testing::quantized_output_scale,
|
|
tflite::testing::quantized_output_zero_point_int8, output_data);
|
|
}
|
|
|
|
TF_LITE_MICRO_TEST(LogisticFloatWideRangeShouldMatchGolden) {
|
|
float output_data[tflite::testing::flat_size_wide_range];
|
|
tflite::testing::TestLogisticFloat(
|
|
tflite::testing::shape_wide_range, tflite::testing::input_data_wide_range,
|
|
tflite::testing::golden_wide_range, tflite::testing::shape_wide_range,
|
|
output_data);
|
|
}
|
|
|
|
TF_LITE_MICRO_TEST(LogisticQuantizedInt8WideRangeShouldMatchGolden) {
|
|
const float input_scale = 1.0;
|
|
const int input_zero_point = 0;
|
|
int8_t input_quantized[tflite::testing::flat_size_wide_range];
|
|
int8_t golden_quantized[tflite::testing::flat_size_wide_range];
|
|
int8_t output_data[tflite::testing::flat_size_wide_range];
|
|
|
|
tflite::testing::TestLogisticQuantized(
|
|
tflite::testing::shape_wide_range, tflite::testing::input_data_wide_range,
|
|
input_quantized, input_scale, input_zero_point,
|
|
tflite::testing::golden_wide_range, golden_quantized,
|
|
tflite::testing::shape_wide_range,
|
|
tflite::testing::quantized_output_scale,
|
|
tflite::testing::quantized_output_zero_point_int8, output_data);
|
|
}
|
|
|
|
TF_LITE_MICRO_TESTS_END
|