numpy 数组的axis

numpy中的axis怎么理解：数组的维数，比如4*5数组，行为axis＝1，列为axis＝0

eg：

a = np.arange(20).reshape(4,5)
print "a:"
print a

print "maximum element in each row of a: " + str(a.max(axis=1))
print "minimum element in each column of a: " + str(a.min(axis=0))

a:

[[ 0 1 2 3 4] [ 5 6 7 8 9] [10 11 12 13 14] [15 16 17 18 19]]

maximum element in each row of a: [ 4 9 14 19]

minimum element in each column of a: [0 1 2 3 4]