What is ndarray in NumPy: Definition and Usage

ndarray is the core data structure in NumPy representing a multidimensional, fixed-size array of elements of the same type. It allows fast and efficient numerical operations on large datasets.

How It Works

An ndarray is like a grid or table that holds numbers or other data types in rows and columns, but it can have more than two dimensions. Imagine a spreadsheet where each cell holds a number; an ndarray is a similar concept but can extend to 3D, 4D, or more.

Under the hood, ndarray stores data in a contiguous block of memory, which makes accessing and changing values very fast. It also keeps track of the shape (how many rows, columns, etc.) and the data type (like integers or floats) so that operations can be done efficiently.

This structure allows NumPy to perform mathematical operations on whole arrays at once, instead of looping through each element, which saves time and makes code simpler.

Example

This example creates a 2D ndarray and shows its shape and data type.

python

import numpy as np

arr = np.array([[1, 2, 3], [4, 5, 6]])
print('Array:')
print(arr)
print('Shape:', arr.shape)
print('Data type:', arr.dtype)

Output

Array:
[[1 2 3]
 [4 5 6]]
Shape: (2, 3)
Data type: int64

When to Use

Use ndarray when you need to work with large amounts of numerical data efficiently. It is ideal for tasks like scientific computing, image processing, machine learning, and any situation where you perform math on arrays of numbers.

For example, if you want to analyze sensor data collected over time or manipulate pixels in an image, ndarray lets you do this quickly and with less code than using regular Python lists.

Key Points

• ndarray is a multidimensional array with fixed size and uniform data type.
• It stores data in a contiguous memory block for fast access.
• Supports vectorized operations for efficient math on whole arrays.
• Essential for numerical and scientific computing in Python.

Key Takeaways

ndarray is the main NumPy structure for fast, multidimensional numerical data.

It stores data in a fixed-size, typed, and contiguous memory block.

Use it to perform efficient math and data manipulation on large datasets.

It simplifies code by enabling operations on whole arrays without loops.