c++ - How can Eigen do dynamic matrix and vector multiplication?

Question

I'm having trouble to do dynamic matrix and vector dot product and surprisingly, I didn't make it find any solution since Eigen is a prevalent library.

So the code is really simple:

int k = 3;
MatrixXd m;
m.resize(k, k);
ArrayXd a;
a.resize(k);
std::cout << "Dot product: " << m*a << std::endl;

I got error

invalid operands to binary expression ('MatrixXd' (aka 'Matrix') and 'ArrayXd' (aka 'Array')) std::cout << "Dot product: " << m*a << std::endl;

I'm confused if doing dynamic matrix and vector multiplication is feasible. Meanwhile, I found that there is .dot() method for vectors and matrices, so which one to use, * or .dot() for dot product?

There is a Matrix-Vector product, but not a Matrix-Array product. Just change the type of a to VectorXd to make your code work. — chtz
I recommend reading this and then this to learn the basics of Eigen. — chtz

Matthieu Brucher Matthieu Brucher · Accepted Answer · 2019-02-01T00:47:45

You need to have matrices, and not a mix of matrices and arrays. You need to convert a to an array (it's a view, no additional computational cost) with .matrix().

Try:

std::cout << "Dot product: " << m*a.matrix() << std::endl;

c++ - How can Eigen do dynamic matrix and vector multiplication?

1 Answers