NumPy Integration in Jupyter

# CHAPTER 11

NumPy Integration in Jupyter #

1. Chapter Introduction #

2. The NumPy Array (Ndarray) #

The core feature of NumPy is the ndarray (N-dimensional array). It looks like a Python list, but it operates entirely differently under the hood.

Cell 1:

import numpy as np
# 'np' is the standard industry alias

# Create a Python list
py_list = [1, 2, 3, 4, 5]

# Convert it to a NumPy array
np_array = np.array(py_list)

print("Python List:", py_list)
print("NumPy Array:", np_array)
print("Type:", type(np_array))

3. Why NumPy Arrays? (Vectorization) #

If you have a list of numbers and want to multiply them all by 2, standard Python requires a slow for loop. NumPy uses Vectorization, allowing you to perform operations on the entire array at once. This is executed in optimized C code, making it incredibly fast.

Cell 2:

# The Slow Python Way
prices_list = [10, 20, 30]
doubled_list = []
for price in prices_list:
    doubled_list.append(price * 2)
print("List Math:", doubled_list)

# The Fast NumPy Way (Vectorization)
prices_array = np.array([10, 20, 30])
doubled_array = prices_array * 2  # Multiplies every element instantly!
print("NumPy Math:", doubled_array)

4. Creating Arrays from Scratch #

NumPy provides built-in functions to quickly generate arrays of numbers without typing them out.

Cell 3:

# 1. Array of zeros (useful for initializing matrices)
zeros = np.zeros(5)
print("Zeros:", zeros)

# 2. Array of ones
ones = np.ones(3)
print("Ones:", ones)

# 3. A sequence of numbers (similar to range())
seq = np.arange(0, 10, 2) # Start at 0, stop at 10, step by 2
print("Sequence:", seq)

# 4. Evenly spaced numbers (useful for plotting graphs)
linspace = np.linspace(0, 1, 5) # 5 numbers evenly spaced between 0 and 1
print("Linspace:", linspace)

5. Multi-Dimensional Arrays (Matrices) #

Machine learning heavily relies on 2D matrices (rows and columns) and 3D tensors.

Cell 4:

# Create a 2D array (Matrix)
matrix = np.array([
    [1, 2, 3],
    [4, 5, 6]
])

print("Matrix:\n", matrix)
print("\nShape (Rows, Cols):", matrix.shape) # Output: (2, 3)

6. Mathematical and Statistical Functions #

NumPy has hundreds of built-in mathematical functions that operate on arrays much faster than the standard Python math library.

Cell 5:

data = np.array([15, 22, 9, 31, 14])

# Statistics
print("Mean:", np.mean(data))
print("Max:", np.max(data))
print("Standard Deviation:", np.std(data))

# Find the INDEX of the maximum value
max_index = np.argmax(data)
print(f"The highest value is at index {max_index}")

7. Mini Project: Performance Testing #

Let's use a Jupyter "Magic Command" (%timeit) to prove why NumPy is the industry standard.

Cell 6:

import numpy as np

# Create 1 million random numbers
massive_list = list(range(1000000))
massive_array = np.arange(1000000)

print("Timing Python List (List Comprehension):")
# %timeit runs the code multiple times to get an accurate average speed
%timeit [x * 2 for x in massive_list]

print("\nTiming NumPy Array (Vectorization):")
%timeit massive_array * 2

*If you run this in Jupyter, you will see that NumPy is often 50x to 100x faster!*

8. Common Mistakes #

• Mixing Data Types: A Python list can hold [1, "Apple", True]. A NumPy array CANNOT. A NumPy array requires every element to be the exact same data type (usually float or int). If you try to mix them, NumPy will convert everything to strings.

• Using math.sqrt() on an array: The standard Python math library does not understand NumPy arrays. You must use np.sqrt(array) instead.

9. MCQs #

10. Interview Questions #

• Q: Explain the concept of Vectorization in NumPy. Why do data scientists use it instead of Python for loops?

• Q: What is the difference between a Python List and a NumPy Array regarding data types?

11. Summary #

12. Next Chapter Recommendation #