You can replace the range over `col` with floats.Scale(1/mu, col), and (minor point) you don't need the underscore in mat.Col
Can I ask why that particular operation? I'm wondering if there's a higher-level way to make your code easier.