Easy NumPy Tutorial Part 2

Introduction

In the previous tutorial we learned about what is NumPy and some basic concepts about ndarrays and how to index them. In this part, we will continue with Boolean Indexing and will also learn about array transposition, universal functions etc.

Boolean Indexing

Let’s consider we have this one dimensional ndarray:

alphas = np.array(['A', 'B', 'A', 'C', 'B', 'A'])

Now assume each element in alphas corresponds to each row in this 2D ndarray:

arr = np.random.randn(6, 4)

Now we want to select those rows which correspond to the alphabet ‘A’. First we will create a boolean array:

alphas == 'A'
# Output: array([ True, False,  True, False, False,  True])

Now we can use this array to get the rows from arr which corresponds to ‘A’:

arr[alphas == 'A']
# Selects rows where alphas is 'A'

Fancy Indexing

You can use integer arrays for indexing, this is known as fancy indexing.

arr = np.arange(24).reshape(6, 4)

# Select particular rows
arr[[1, 5, 3, 0]]
# Output: array([[ 4,  5,  6,  7],
#                [20, 21, 22, 23],
#                [12, 13, 14, 15],
#                [ 0,  1,  2,  3]])

Array Transposition

You can compute array transposition with the special .T attribute:

arr = np.random.randn(3, 4)
arr.T
# Transposes the array

Universal Functions (ufuncs)

ufuncs are functions that perform element wise operations on data present in ndarrays:

ar = np.arange(10)
np.sqrt(ar)
# Output: array([0., 1., 1.41421356, 1.73205081, 2., ...])

Using np.where

Suppose we have two ndarrays x and y, and we want to choose a value from x when the value is greater than 5 otherwise choose from y:

x = np.array([4, 6, 7, 4, 6, 4, 2])
y = np.array([8, 4, 7, 9, 1, 3, 6])
np.where(x > 5, x, y)
# Output: array([8, 6, 7, 9, 6, 3, 6])

Replace all negative values with zero:

arr = np.random.randn(3, 4)
np.where(arr > 0, arr, 0)

Mathematical and Statistical Methods

NumPy provides statistical and mathematical methods like sum, mean, std etc:

arr = np.arange(20).reshape(5, 4)

np.sum(arr)
# Output: 190

np.mean(arr)
# Output: 9.5

# Sum across columns
arr.sum(axis=1)
# Output: array([ 6, 22, 38, 54, 70])

Boolean Array Methods

The method sum counts True values:

arr = np.array([True, True, False, True, True])
arr.sum()
# Output: 4

arr.any()  # Returns True if any True exists
# Output: True

arr.all()  # Returns True if all are True
# Output: False

Continue to Part 3 for more NumPy concepts!