How to Round Numbers in Python?

Last Updated : 30 Apr, 2025

Rounding a number simplifies it while keeping its value as close as possible to the original. Python provides various methods to round numbers, depending on how we want to handle the precision or rounding behavior. In this article, we'll cover the most commonly used techniques for rounding numbers in Python, along with examples. For example, Input is 3.5 then Output should be 4.

Using Built-in round() Function

Python’s built-in round() function rounds a number to a given precision. If no precision is specified, it rounds to the nearest integer.

python

print(round(11))
print(round(22.7))  
print(round(4.465, 2)) 
print(round(4.476, 2))   
print(round(4.473, 2))

Output

Explanation:

round(x) rounds x to the nearest integer.
round(x, n) rounds x to n decimal places.

Using Truncation (Cutting Off Digits)

Truncation cuts off digits after a specified number of decimal places without rounding. It works for both positive and negative numbers.

python

def truncate(num, dec=0):
    mult = 10 ** dec
    return int(num * mult) / mult

print(truncate(16.5))
print(truncate(-3.853, 1))
print(truncate(3.815, 2))
print(truncate(346.8, -1))
print(truncate(-2947.48, -3))

Output

16.0
-3.8
3.81
340.0
-2000.0

Explanation:

Multiplies by 10^dec, truncates, and divides back.
Useful when you only want to chop off digits, not round.

Using math.ceil() and math.floor()

Python’s math module provides two essential functions:

ceil(x): Rounds up to the nearest integer.
floor(x): Rounds down to the nearest integer.

python

import math

print(math.ceil(4.2))
print(math.ceil(-0.5))
print(math.floor(2.2))
print(math.floor(-0.5))

Output

Explanation:

ceil(x) always rounds up.
floor(x) always rounds down, even for negative numbers.

Round Up to Specific Decimal Places

If we want to round up to a specified decimal place, you can combine multiplication, math.ceil(), and division.

python

import math

def round_up(num, dec=0):
    mult = 10 ** dec
    return math.ceil(num * mult) / mult

print(round_up(2.1))
print(round_up(2.23, 1))
print(round_up(2.543, 2))
print(round_up(22.45, -1))
print(round_up(2352, -2))

Output

3.0
2.3
2.55
30.0
2400.0

Explanation:

Shifts decimal right, applies ceil(), and shifts back.
Works for positive and negative numbers.

We can follow the diagram below to understand round up and round down. Round up to the right and down to the left.

Understanding Rounding Up and Rounding Down

Rounding up always rounds a number to the right on the number line and rounding down always rounds a number to the left on the number line.

Round Down to Specific Decimal Places

Similarly, for rounding down we can use floor() function instead of ceil():

python

import math

def round_down(num, dec=0):
    mult = 10 ** dec
    return math.floor(num * mult) / mult

print(round_down(2.5))
print(round_down(2.48, 1))
print(round_down(-0.5))

Output

2.0
2.4
-1.0

Explanation: Same shifting logic as round up, but using floor().

Round Numbers Using NumPy

NumPy provides the np.round() function for rounding arrays or numbers efficiently.

Python

import numpy as np

arr = np.array([1.234, 2.567, 3.789])

print("Nearest integer:", np.round(arr))
print("Two decimals:", np.round(arr, 2))

Output

Nearest integer: [1. 3. 4.]
Two decimals: [1.23 2.57 3.79]

Explanation:

np.round(arr) rounds to the nearest whole number.
np.round(arr, n) rounds to n decimals.

Round Numbers in Python using Rounding Bias concept

When rounding large datasets, we must avoid rounding bias. Python supports several rounding methods for that and most generally used methods are:

1. Rounding Half Up

Always rounds .5 upward.

python

import math

def round_half_up(num, dec=0):
    mult = 10 ** dec
    return math.floor(num * mult + 0.5) / mult

print(round_half_up(1.28, 1))
print(round_half_up(-1.5))
print(round_half_up(-1.225, 2))

Output

1.3
-1.0
-1.23

Explanation: Adds 0.5 before flooring to simulate "round half up."

2. Rounding Half Down

Always rounds .5 downward.

python

import math

def round_half_down(num, dec=0):
    mult = 10 ** dec
    return math.ceil(num * mult - 0.5) / mult

print(round_half_down(2.5))
print(round_half_down(-2.5))
print(round_half_down(2.25, 1))

Output

2.0
-3.0
2.2

Explanation: Subtracts 0.5 before applying ceil() to simulate "round half down."

Rounding Half to Even (Bankers' Rounding)

Python’s built-in round() uses round half to even strategy, which rounds .5 to the nearest even number to minimize bias.

If n is positive and d >= 5, round up
If n is positive and d = 5, round down
If n is negative and d >= 5, round down
If n is negative and d < 5, round up

After rounding as per the rules mentioned above, we can shift the decimal place back to the left.

python

from decimal import Decimal

print(Decimal("0.1"))
print(Decimal(0.1))
print(Decimal("1.65").quantize(Decimal("1.0")))
print(Decimal("1.675").quantize(Decimal("1.00")))

Output