1. What is a NumPy Array?

A NumPy array (ndarray) is the core data structure in NumPy.

  • Stands for n-dimensional array
  • Represents data as vectors
  • Supports fast, efficient computation

Key Point:
ndarray = foundation of NumPy

2. Why Arrays are Powerful

Arrays enable:

  • Vectorization → operations on all elements at once
  • Faster computation
  • Lower memory usage

Key Point:
Arrays process data efficiently

3. Creating an Array

Convert Python objects (like lists) into arrays:

np.array([1, 2, 3, 4])

Key Point:
Use np.array() to create arrays

4. Mutability of Arrays

Arrays are mutable:

  • You can change element values

Example:

arr[-1] = 5

Key Point:
Values can be modified

5. Array Size Limitation

  • Cannot change size directly
  • Must reassign to modify size

Example:

  • Cannot append like lists

Key Point:
Size change requires reassignment

6. Data Type Consistency

All elements must have the same data type.

Example:

  • Mixing int + string → all become string

Key Point:
Arrays enforce uniform data types

7. Checking Array Type

Use:

type(arr)

Returns:

  • NumPy array type

Key Point:
Array is a specific object type

8. Checking Data Type (dtype)

arr.dtype

Returns:

  • Data type of elements

Example:

  • int, float, U21 (string)

Key Point:
dtype shows element type

9. Dimensions of Arrays

1D Array

  • Single sequence
  • Shape: (n,)

Not row or column

2D Array

  • List of lists
  • Like a table

Example:

  • Rows × Columns

3D Array

  • List of 2D arrays
  • Like multiple tables

Key Point:
More dimensions = more complexity

10. Shape of Array

Use:

arr.shape

Returns:

  • Dimensions (rows, columns)

Example:

  • (4, 2) → 4 rows, 2 columns

Key Point:
Shape describes structure

11. Number of Dimensions (ndim)

Use:

arr.ndim

Returns:

  • Number of dimensions

Example:

  • 2 → 2D array

Key Point:
ndim = dimensionality

12. Reshaping Arrays

Change structure using:

arr.reshape(new_shape)

Example:

  • (4,2) → (2,4)

Must reassign:

arr = arr.reshape(2,4)

Key Point:
Reshape reorganizes data

13. Mathematical Operations

NumPy provides built-in functions:

  • Mean → np.mean()
  • Log → np.log()
  • Floor → round down
  • Ceiling → round up

Key Point:
NumPy simplifies math operations

14. Practical Importance

Arrays are used for:

  • Data transformation
  • Statistical analysis
  • Machine learning preprocessing

Key Point:
Core structure for numerical data

15. Relationship with Pandas

  • Pandas is built on NumPy
  • Understanding NumPy helps:
    • DataFrames
    • Data analysis

Key Point:
NumPy = foundation of pandas

16. Debugging Insight

Use:

  • .shape
  • .ndim
  • .dtype

Helps:

  • Understand errors
  • Fix mismatches

Key Point:
Inspect arrays when debugging

17. Practical Limitation

  • Usually work with:
    • 1D, 2D, or 3D arrays
  • Higher dimensions are rare

Key Point:
Most real-world data is 2D

Final Summary

NumPy arrays (ndarray) are powerful data structures designed for efficient numerical computation. They support vectorized operations, enforce consistent data types, and can represent data in multiple dimensions. Arrays are mutable but have fixed sizes, requiring reassignment for structural changes. Attributes like shape, ndim, and dtype help understand and debug arrays. NumPy forms the foundation for many data science tools, including pandas, making it essential for data professionals.

Key Takeaways

  • ndarray = core NumPy structure
  • Supports vectorized operations
  • Faster than lists
  • Same data type required
  • Mutable but fixed size
  • Use .shape, .ndim, .dtype
  • Use reshape() to change structure
  • Foundation for pandas
  • Essential for data analysis