python - Finding unique points in numpy array

Question

What is a faster way of finding unique x,y points (removing duplicates) in a numpy array like:

points = numpy.random.randint(0, 5, (10,2))

I thought of converting points to a complex numbers and then checking for unique, but that seems rather convoluted:

b = numpy.unique(points[:,0] + 1j * points[:,1])
points = numpy.column_stack((b.real, b.imag))

Answer 1

I would do it like this:

numpy.array(list(set(tuple(p) for p in points)))

For the fast solution in the most general case, maybe this recipe would interest you: http://code.activestate.com/recipes/52560-remove-duplicates-from-a-sequence/