Basic Operations and Advanced Features of NumPy

NumPy goes beyond simple array operations to offer a variety of essential functions in data analysis and machine learning.

In this lesson, we'll delve into basic array operations, broadcasting, and random number generation.

NumPy arrays support basic arithmetic operations, and these operations are carried out as follows:

Copy
import numpy as np

arr1 = np.array([1, 2, 3])
arr2 = np.array([4, 5, 6])

print(arr1 + arr2)  # [5 7 9]
print(arr1 * arr2)  # [ 4 10 18]
print(arr1 - arr2)  # [-3 -3 -3]
print(arr1 / arr2)  # [0.25 0.4  0.5 ]

Unlike Python lists, NumPy arrays automatically perform element-wise operations.

NumPy makes it simple to change the shape of an array.

Copy
arr = np.array([1, 2, 3, 4, 5, 6])

# Convert to 2x3 matrix
reshaped_arr = arr.reshape(2, 3)

print(reshaped_arr)

The ability to dynamically resize arrays aids significantly in data preprocessing.

In NumPy arrays, specific elements can be selected just like with lists.

Copy
arr = np.array([10, 20, 30, 40, 50])

print(arr[0])  # 10
print(arr[1:4])  # [20 30 40]

Elements in multi-dimensional arrays can also be specifically selected by row and column.

Copy
matrix = np.array([[1, 2, 3], [4, 5, 6]])

print(matrix[0, 1])  # 2 (second element in the first row)
print(matrix[:, 1])  # [2 5] (second column of every row)

Broadcasting is a key NumPy feature that enables operations between arrays of different sizes.

Copy
arr1 = np.array([[1, 2, 3], [4, 5, 6]])
arr2 = np.array([10, 20, 30])

# `arr2` is automatically expanded across each row
result = arr1 + arr2

print(result)

In the above example, although arr2 has a shape of (1,3), NumPy automatically expands it to (2,3) for the operation.

Thus, even when array sizes don't match, broadcasting allows for efficient operations.

NumPy makes it easy to extract data using conditions.

Copy
arr = np.array([10, 20, 30, 40, 50])

# Select values greater than 30
filtered = arr[arr > 30]

print(filtered)

NumPy provides random number generation capabilities, useful for data sampling and simulations.

Copy
# Generate 5 random numbers between 0 and 1
random_arr = np.random.rand(5)

print(random_arr)

NumPy is a versatile library essential for AI and data science.

You can swiftly create arrays, perform operations, and apply mathematical computations with ease.

• NumPy Official Documentation