Time for action – fancy indexing in-place for ufuncs with the at() method

To demonstrate how the at() method works, start a Python or IPython shell and import NumPy. You should know how to do this by now.

1. Create an array with seven random integers from -3 to 3 with a seed of 42:

   >>> a = np.random.random_integers(-3, 3, 7)
   >>> a
   array([ 1,  0, -1,  2,  1, -2,  0])
   

   When we talk about random numbers in programming, we usually talk about pseudo-random numbers (see https://www.khanacademy.org/computing/computer-science/cryptography/crypt/v/random-vs-pseudorandom-number-generators). The numbers appear random, but in fact are calculated using a seed.

2. Apply the at() method of the sign() universal function to the fourth and sixth array elements:

   >>> np.sign.at(a, [3, 5])
   >>> a
   array([ 1, 0, -1,  1,  1, -1,  0])
   

What just happened?

We used the at() method to select array elements and performed an in-place operation—determining the sign. We also learned...