TensorTests.cpp

Go to the documentation of this file.

#include <gtest/gtest.h>

#include <fstream>
// Include headers that implement a archive in simple text format
#include <boost/archive/text_iarchive.hpp>
#include <boost/archive/text_oarchive.hpp>

// Redefine word "public" so every class field/method will be accessible for tests.
#define private public
#include <types/Tensor.hpp>
#include <types/Matrix.hpp>

TEST(Tensor, Dimensions2x5) {
    // Default sizes of matrices.
    const size_t N = 2;
    const size_t M = 5;
    const size_t K = 13;

    mic::types::Tensor<float> nm({N, M, K});

    ASSERT_EQ(nm.dim(0), N);
    ASSERT_EQ(nm.dim(1), M);
    ASSERT_EQ(nm.dim(2), K);
}


TEST(Tensor, Enumeration2x3) {
    // Default sizes of matrices.
    const size_t N = 2;
    const size_t M = 3;

    mic::types::Tensor<float> nm({N, M});
    nm.enumerate();

//  std::cout << nm << std::endl;

    for (size_t i =0; i< N*M; i++)
        ASSERT_EQ(nm(i), i);

/*  for(size_t row=0; row<N; row++) {
        for(size_t col=0; col<M; col++)
            std::cout << " nm(" << row << "," << col << ") = " << nm({row,col});
        std::cout << std::endl;
    }//: for*/

    ASSERT_EQ(nm({0,0}), 0);
    ASSERT_EQ(nm({1,0}), 1);
    ASSERT_EQ(nm({0,1}), 2);
    ASSERT_EQ(nm({1,1}), 3);
    ASSERT_EQ(nm({0,2}), 4);
    ASSERT_EQ(nm({1,2}), 5);

}


TEST(Tensor, Serialization) {
    // Default sizes of matrices.
    const size_t N = 2;
    const size_t M = 5;
    const size_t K = 11;

    mic::types::Tensor<float> nm({N, M, K});
    nm.randn();

    const char* fileName = "saved.txt";
    // Save data
    {
        // Create an output archive
        std::ofstream ofs(fileName);
        boost::archive::text_oarchive ar(ofs);
        // Write data
        ar & nm;
        //std::cout << "Saved matrix = " << nm << std::endl;
    }

    // Restore data
    mic::types::Tensor<float> restored_tensor;
    restored_tensor.randn();

    {
        // Create and input archive
        std::ifstream ifs(fileName);
        boost::archive::text_iarchive ar(ifs);
        // Load data
        ar & restored_tensor;
        //std::cout << "Restored tensor = " << restored_tensor << std::endl;
    }

    // Check dimensions.
    ASSERT_EQ(nm.elements, restored_tensor.elements);
    ASSERT_EQ(nm.dimensions.size(), restored_tensor.dimensions.size());
    for (size_t i =0; i< (size_t)nm.dimensions.size(); i++)
        ASSERT_EQ(nm.dimensions[i], restored_tensor.dimensions[i]);

    // Check values.
    float eps = 1e-8;
    for (size_t i =0; i< (size_t)nm.size(); i++)
        EXPECT_LE(fabs(nm(i) - restored_tensor(i)), eps);

}

TEST(Tensor, Im2Col2x3) {
    // Default sizes of matrices.
    const size_t N = 2;
    const size_t M = 3;

    mic::types::Tensor<float> nm({N, M});
    nm.enumerate();


}

int main(int argc, char **argv) {
    testing::InitGoogleTest(&argc, argv);
    return RUN_ALL_TESTS();
}