Run this notebook: Open in Colab Open in Kaggle

import numpy as np

Array Creation Methods

NumPy provides multiple ways to create arrays beyond np.array(). np.arange() generates evenly spaced integers (like Python’s range() but returning an array). np.zeros() and np.full() create arrays pre-filled with a specific value – zeros are commonly used as accumulators or placeholder tensors, while full() lets you initialize with any constant. You can also convert existing Python lists to NumPy arrays with np.array(). Understanding these factory functions lets you quickly set up test data, initialize model weights, and prepare data structures for computation.

# Create a numpy array
np1 = np.array([0,1,2,3,4,5,6,7,8,9])
print(np1)
print(np1.shape)

# Arange
np2 = np.arange(10)
print(np2)

# step
np3 = np.arange(0,10, 2)
print(np3)

# Zeros
np4 = np.zeros(10)
print(np4)

# Multidimensional zeros
np5 = np.zeros((2,10))
print(np5)

# Full
np6 = np.full((10), 6)
print(np6)

# Multidimensional Full
np7 = np.full((2,10), 6)
print(np7)

# Convert a python list to a np
my_list3 = [0,1,2,3,4,5,6,7,8,9]
print(my_list3)
np8 = np.array(my_list3)
print(np8)
print(np8[0])