NumPy ndarray.nbytes Attribute

NumPy ndarray.nbytes

The ndarray.nbytes attribute returns the total number of bytes consumed by the elements of a NumPy array. It does not include memory used for metadata such as shape, dtype, or strides.

Syntax

ndarray.nbytes

Return Value

Type	Description
int	Returns the total number of bytes consumed by the elements of the array.

---

Examples

1. Checking Memory Usage of an Integer Array

This example demonstrates how nbytes calculates the total memory usage of an array.

import numpy as np

# Creating a NumPy array of integers
arr = np.array([1, 2, 3, 4, 5], dtype=np.int32)

# Getting total bytes consumed by the elements
bytes_used = arr.nbytes
print("Total bytes used by array elements:", bytes_used)

Output:

Total bytes used by array elements: 20

Since each integer (int32) consumes 4 bytes and there are 5 elements, the total memory usage is 5 × 4 = 20 bytes.

2. Checking Memory Usage of a Float Array

Here, we check how much memory a float array consumes.

import numpy as np

# Creating a NumPy array of floats
arr = np.array([1.1, 2.2, 3.3, 4.4], dtype=np.float64)

# Getting total bytes consumed by the elements
bytes_used = arr.nbytes
print("Total bytes used by array elements:", bytes_used)

Output:

Total bytes used by array elements: 32

Since each float64 element occupies 8 bytes and there are 4 elements, the total memory usage is 4 × 8 = 32 bytes.

3. Memory Usage of a Multi-Dimensional Array

In this example, we examine the memory consumption of a 2D array.

import numpy as np

# Creating a 2D NumPy array of integers
arr = np.array([[10, 20, 30], 
                [40, 50, 60]], dtype=np.int16)

# Getting total bytes consumed by the elements
bytes_used = arr.nbytes
print("Total bytes used by array elements:", bytes_used)

Output:

Total bytes used by array elements: 12

Each int16 value takes 2 bytes. The array has 6 elements, so the total memory usage is 6 × 2 = 12 bytes.

4. Comparing Memory Usage of Different Data Types

This example shows how different data types impact memory consumption.

import numpy as np

# Creating NumPy arrays with different data types
arr_int8 = np.array([1, 2, 3, 4], dtype=np.int8)
arr_int64 = np.array([1, 2, 3, 4], dtype=np.int64)
arr_float32 = np.array([1.0, 2.0, 3.0, 4.0], dtype=np.float32)

# Checking memory usage for each array
print("Bytes used by int8 array:", arr_int8.nbytes)
print("Bytes used by int64 array:", arr_int64.nbytes)
print("Bytes used by float32 array:", arr_float32.nbytes)

Output:

Bytes used by int8 array: 4
Bytes used by int64 array: 32
Bytes used by float32 array: 16

The memory usage differs because:

int8 uses 1 byte per element (4 × 1 = 4 bytes).
int64 uses 8 bytes per element (4 × 8 = 32 bytes).
float32 uses 4 bytes per element (4 × 4 = 16 bytes).