/********************************************************************************
* Copyright (c) 2023 CEA-List
*
* This program and the accompanying materials are made available under the
* terms of the Eclipse Public License 2.0 which is available at
* http://www.eclipse.org/legal/epl-2.0.
*
* SPDX-License-Identifier: EPL-2.0
*
********************************************************************************/
#include <catch2/catch_test_macros.hpp>
#include <cstddef> // std::size_t
#include <cstdint> // std::uint16_t
#include <chrono>
#include <iostream>
#include <memory>
#include <numeric> // std::accumulate
#include <random> // std::random_device, std::mt19937, std::uniform_real_distribution
#include "aidge/data/Tensor.hpp"
#include "aidge/backend/cpu/data/TensorImpl.hpp"
#include "aidge/operator/Add.hpp"
#include "aidge/backend/cpu/operator/AddImpl.hpp"
namespace Aidge {
TEST_CASE("Test addition of Tensors","[TensorImpl][Add]") {
constexpr std::uint16_t NBTRIALS = 10;
// Create a random number generator
std::random_device rd;
std::mt19937 gen(rd());
std::uniform_real_distribution<float> valueDist(0.1f, 1.1f); // Random float distribution between 0 and 1
std::uniform_int_distribution<std::size_t> dimSizeDist(std::size_t(2), std::size_t(10));
std::uniform_int_distribution<int> boolDist(0,1);
// Create MatMul Operator
std::shared_ptr<Node> mySub = Add(2);
auto op = std::static_pointer_cast<OperatorTensor>(mySub-> getOperator());
op->setDataType(DataType::Float32);
op->setBackend("cpu");
// Create 2 input Tensors
std::shared_ptr<Tensor> T0 = std::make_shared<Tensor>();
op->associateInput(0,T0);
T0->setDataType(DataType::Float32);
T0->setBackend("cpu");
std::shared_ptr<Tensor> T1 = std::make_shared<Tensor>();
op -> associateInput(1,T1);
T1->setDataType(DataType::Float32);
T1->setBackend("cpu");
// Create results Tensor
Tensor Tres{};
Tres.setDataType(DataType::Float32);
Tres.setBackend("cpu");
// To measure execution time of 'MatMul_Op::forward()' member function call
std::chrono::time_point<std::chrono::system_clock> start;
std::chrono::time_point<std::chrono::system_clock> end;
std::chrono::duration<double, std::micro> duration{};
std::size_t number_of_operation = 0;
for (std::uint16_t trial = 0; trial < NBTRIALS; ++trial) {
// generate 2 random Tensors
// handle dimensions, replace some dimensions with '1' to get broadcasting
constexpr std::size_t nbDims = 4;
std::vector<std::size_t> dims;
for (std::size_t i = 0; i < nbDims; ++i) {
dims.push_back(dimSizeDist(gen));
}
std::vector<std::size_t> dims0 = dims;
std::vector<std::size_t> dims1 = dims;
std::vector<std::size_t> dimsOut = dims;
for (std::size_t i = 0; i < nbDims; ++i) {
if (boolDist(gen)) {
dims0[i] = 1;
}
if (boolDist(gen)) {
dims1[i] = 1;
}
dimsOut[i] = (dims0[i] == 1) ? dims1[i] : dims0[i];
}
// create arrays and fill them with random values
float* array0 = new float[dims0[0]*dims0[1]*dims0[2]*dims0[3]];
float* array1 = new float[dims1[0]*dims1[1]*dims1[2]*dims1[3]];
float* result = new float[dimsOut[0]*dimsOut[1]*dimsOut[2]*dimsOut[3]];
for (std::size_t i = 0; i < dims0[0]*dims0[1]*dims0[2]*dims0[3]; ++i) {
array0[i] = valueDist(gen);
}
for (std::size_t i = 0; i < dims1[0]*dims1[1]*dims1[2]*dims1[3]; ++i) {
array1[i] = valueDist(gen);
}
// compute true result
const std::size_t strides0[nbDims] = {dims0[1]*dims0[2]*dims0[3], dims0[2]*dims0[3], dims0[3], 1};
const std::size_t strides1[nbDims] = {dims1[1]*dims1[2]*dims1[3], dims1[2]*dims1[3], dims1[3], 1};
for (std::size_t a = 0; a < dimsOut[0]; ++a) {
for (std::size_t b = 0; b < dimsOut[1]; ++b) {
const std::size_t idx0_0 = strides0[0] * ((dims0[0] > 1) ? a : 0)
+ strides0[1] * ((dims0[1] > 1) ? b : 0);
const std::size_t idx1_0 = strides1[0] * ((dims1[0] > 1) ? a : 0)
+ strides1[1] * ((dims1[1] > 1) ? b : 0);
for (std::size_t c = 0; c < dimsOut[2]; ++c) {
const std::size_t idx_out = dimsOut[3] * (c + dimsOut[2] * (b + dimsOut[1] * a));
for (std::size_t d = 0; d < dimsOut[3]; ++d) {
std::size_t idx0 = idx0_0
+ strides0[2] * ((dims0[2] > 1) ? c : 0)
+ ((dims0[3] > 1) ? d : 0);
std::size_t idx1 = idx1_0
+ strides1[2] * ((dims1[2] > 1) ? c : 0)
+ ((dims1[3] > 1) ? d : 0);
result[idx_out + d] = array0[idx0] + array1[idx1];
// std::cout << "(" << idx0 << ", " << idx1 << ") -> " << array0[idx0] << " - " << array1[idx1] << " -> " << idx_out + d << std::endl;
}
}
}
}
// conversion to Aidge::Tensors
// input0
T0->resize(dims0);
T0->getImpl() -> setRawPtr(array0, dims0[0]*dims0[1]*dims0[2]*dims0[3]);
// input1
T1->resize(dims1);
T1->getImpl() -> setRawPtr(array1, dims1[0]*dims1[1]*dims1[2]*dims1[3]);
// results
Tres.resize(dimsOut);
Tres.getImpl() -> setRawPtr(result, dimsOut[0]*dimsOut[1]*dimsOut[2]*dimsOut[3]);
Tensor T2 = *T0 + *T1;
REQUIRE(T2 == Tres);
// no implementation
Tensor T3(T1->dims());
REQUIRE_THROWS(*T0 + T3);
// // wrong backend
// static Registrar<Add_Op> registrarAddImpl_custom("custom", [](const Add_Op& op) { return std::make_unique<AddImpl_cpu>(op); } );
// static Registrar<Tensor> registrarTensorImpl_custom_Int32({"custom", DataType::Int32},
// [] (DeviceIdx_t device, std::vector<DimSize_t> dims) {
// return std::make_shared<TensorImpl_cpu<int>>(device, dims);
// }
// );
// T1.setBackend("custom");
// REQUIRE_THROWS(T0 + T1);
// wrong datatype
Tensor T4(T1->dims());
T4.setDataType(DataType::Float64);
REQUIRE_THROWS(*T0 + T4);
}
}
TEST_CASE("Test substraction of Tensors","[TensorImpl][Sub]") {
Tensor T0 = Array3D<int, 2, 2, 2>{{{{1, 2}, {3, 4}}, {{5, 6}, {7, 8}}}};
Tensor T1 = Array3D<int, 2, 2, 2>{{{{7, 1}, {3, 7}}, {{54, 0}, {7, 12}}}};
Tensor T2 = T0 - T1;
T2.print();
REQUIRE(T2 == Tensor(Array3D<int, 2, 2, 2>{{{{-6,1},{0,-3}},{{-49,6},{0,-4}}}}));
Tensor T3(T1.dims());
REQUIRE_THROWS(T0 - T3);
}
TEST_CASE("Test multiplication of Tensors","[TensorImpl][Mul]") {
Tensor T0 = Array3D<int, 2, 2, 2>{{{{1, 2}, {3, 4}}, {{5, 6}, {7, 8}}}};
Tensor T1 = Array3D<int, 2, 2, 2>{{{{7, 2}, {3, 7}}, {{5, 6}, {7, 8}}}};
Tensor T2 = T0 * T1;
T2.print();
REQUIRE(T2 == Tensor(Array3D<int, 2, 2, 2>{{{{7,4},{9,28}},{{25,36},{49,64}}}}));
Tensor T3(T1.dims());
REQUIRE_THROWS(T0 * T3);
}
TEST_CASE("Test division of Tensors","[TensorImpl][Div]") {
Tensor T0 = Array3D<int, 2, 2, 2>{{{{7,4},{9,28}},{{25,36},{49,64}}}};
Tensor T1 = Array3D<int, 2, 2, 2>{{{{7, 2}, {3, 7}}, {{5, 6}, {7, 8}}}};
Tensor T2 = T0 / T1;
T2.print();
REQUIRE(T2 == Tensor(Array3D<int, 2, 2, 2>{{{{1, 2}, {3, 4}}, {{5, 6}, {7, 8}}}}));
Tensor T3(T1.dims());
REQUIRE_THROWS(T0 / T3);
}
} // namespace Aidge