Matrices using vectors

Vectors can be used as a 2D matrix by defining them as a vector of vectors.

A matrix with 3 rows and 4 columns with each cell initialised as 0 can be defined as:

std::vector<std::vector<int> > matrix(3, std::vector<int>(4));

The syntax for initializing them using initialiser lists or otherwise are similar to that of a normal vector.

std::vector<std::vector<int>> matrix = { {0,1,2,3},
                                         {4,5,6,7}, 
                                         {8,9,10,11}
                                       };

Values in such a vector can be accessed similar to a 2D array

int var = matrix[0][2];

Iterating over the entire matrix is similar to that of a normal vector but with an extra dimension.

for (int i = 0; i < 3; ++i)
{
    for(int j = 0; j < 4; ++j)
    {
        std::cout << matrix[i][j] << std::endl;
    }
}

for (auto& row: matrix)
{
    for(auto& col : row)
    { 
        std::cout << col << std::endl;
    }
}

A vector of vectors is a convenient way to represent a matrix but it’s not the most efficient: individual vectors are scattered around memory and the data structure isn’t cache friendly.

Also, in a proper matrix, the length of every row must be the same (this isn’t the case for a vector of vectors). The additional flexibility can be a source of errors.