Given the following NumPy array,

> a = array([[1, 2, 3, 4, 5], [1, 2, 3, 4, 5],[1, 2, 3, 4, 5]])

it's simple enough to shuffle a single row,

> shuffle(a[0])
> a
array([[4, 2, 1, 3, 5],[1, 2, 3, 4, 5],[1, 2, 3, 4, 5]])

Is it possible to use indexing notation to shuffle each of the rows independently? Or do you have to iterate over the array. I had in mind something like,

> numpy.shuffle(a[:])
> a
array([[4, 2, 3, 5, 1],[3, 1, 4, 5, 2],[4, 2, 1, 3, 5]]) # Not the real output

though this clearly doesn't work.

You have to call numpy.random.shuffle() several times because you are shuffling several sequences independently. numpy.random.shuffle() works on any mutable sequence and is not actually a ufunc. The shortest and most efficient code to shuffle all rows of a two-dimensional array a separately probably is

list(map(numpy.random.shuffle, a))

Some people prefer to write this as a list comprehension instead:

[numpy.random.shuffle(x) for x in a]