No, there's no such way.
The problem is that this kind of thing doesn't compose. There are many kinds of operations one might want to do (sum, max, mean, std, etc.). Instead, use mat.Row and floats (or appropriate changes for other functions).
sums := make([]float64, c)
col := make([]float64, r)
for i := 0; i < r ; i++{
mat.Col(col, a)
sums[i] = floats.Sum(col)
}
Note that if you have a Dense matrix, you can also use dense.RawRowView to speed things up