Sorting arrays in NumPy by column

Question

How can I sort an array in NumPy by the nth column?

For example,

a = array([[9, 2, 3],
           [4, 5, 6],
           [7, 0, 5]])

Answer 1

A little more complicated lexsort example - descending on the 1st column, secondarily ascending on the 2nd. The tricks with lexsort are that it sorts on rows (hence the .T), and gives priority to the last.

In [120]: b=np.array([[1,2,1],[3,1,2],[1,1,3],[2,3,4],[3,2,5],[2,1,6]])
In [121]: b
Out[121]: 
array([[1, 2, 1],
       [3, 1, 2],
       [1, 1, 3],
       [2, 3, 4],
       [3, 2, 5],
       [2, 1, 6]])
In [122]: b[np.lexsort(([1,-1]*b[:,[1,0]]).T)]
Out[122]: 
array([[3, 1, 2],
       [3, 2, 5],
       [2, 1, 6],
       [2, 3, 4],
       [1, 1, 3],
       [1, 2, 1]])