How to solve representation errors with floating-point numbers in Python?

Question

I need to truncate a decimal at the hundredths place without rounding. I'm using the the method:

y = 78.459

x = int(y * 100) / 100

which comes back as 78.45. As it should.

However, if I do

y = 5.10

x = int(y * 100) / 100

it come back as 5.09, when it should be 5.10

I know that this is a representation error, due to the floating-point inaccuracies in Python, but I don't know how to fix it. If anyone can either help me figure out how to fix representation errors or maybe find a better way to truncate at the hundredths place, I would really appreciate it. Thanks

Check this out: https://stackoverflow.com/questions/11522933/python-floating-point-arbitrary-precision-available — Trauer, Sep 18 '18 at 01:20

U13-Forward · Accepted Answer · 2018-09-18T01:26:10.430

1

Use decimal module for this:

>>> from decimal import Decimal as d
>>> y = d('5.10')
>>> x = int(y * 100) / 100
>>> x
5.1
>>>

edited Sep 18 '18 at 01:26

answered Sep 18 '18 at 01:20

U13-Forward

69,221
14
89
114

This is not really the proper use of Decimal. – Stephen Rauch Sep 18 '18 at 01:29
@StephenRauch Really?, then what's the best then – U13-Forward Sep 18 '18 at 01:30
1

Okay, thanks. Can I pass a value to d() from input()? – Hollyleaf25 Sep 18 '18 at 01:45
@Hollyleaf25 What you mean? – U13-Forward Sep 18 '18 at 01:45
Like if x = input("enter a number: ") then could I do y = d(x)? – Hollyleaf25 Sep 18 '18 at 01:46
@Hollyleaf25 Can pass `d` to any float string like i.e (`"12.35"`,`"13.34"`, `"2435.54363"`) – U13-Forward Sep 18 '18 at 01:46

score 0 · Answer 2 · answered Sep 19 '18 at 16:47

0

You can use the quantize function in the decimal module:

>>> import decimal
>>> y = decimal.Decimal('78.459')
>>> z = decimal.Decimal('.01')
>>> y.quantize(z, rounding=decimal.ROUND_DOWN)
Decimal('78.45')

answered Sep 19 '18 at 16:47

Simon Byrne

7,694
1
26
50

How to solve representation errors with floating-point numbers in Python?

2 Answers2