Replace zero in the matrix by 1e-15 to avoid infinite values when CLR transformation is computed

Replace zero in the matrix by 1e-15 to avoid infinite values when CLR transformation is computed

Usage

replace_zero(x)