+- Python Forum (https://python-forum.io)
+-- Forum: Python Coding (https://python-forum.io/forum-7.html)
+--- Forum: General Coding Help (https://python-forum.io/forum-8.html)
+--- Thread: python calculate float plus float is incorrect? (/thread-41967.html)
python calculate float plus float is incorrect? - sirocawa - Apr-16-2024
hi everybody
maybe , try print(27.56+13.78)
I use 3.10.12 and 3.12, get result is 41.339999999999996.
what happend?
RE: python calculate float plus float is incorrect? - Gribouillis - Apr-16-2024
It is all explained here. Also it has nothing to do with Python. It's due to the internal representation of floating point numbers by computers. The same issue exists in other programming languages.
RE: python calculate float plus float is incorrect? - sirocawa - Apr-16-2024
but, i use php to calcu, the result is correct 41.34
<?php
$a=27.56;
$b=13.78;
echo $a+$b;
?>
41.34
RE: python calculate float plus float is incorrect? - Gribouillis - Apr-16-2024
I don't know PHP, but here is a warning about the same thing in PHP. I don't know why it displays 41.34 in your case (perhaps an effect of the echo command?). Try their example perhaps
Output:floor((0.1+0.7)*10) RE: python calculate float plus float is incorrect? - Axel_Erfurt - Apr-16-2024
a = 27.56
b = 13.78
c = a + b
print(f"{c:.2f}")Output:41.34RE: python calculate float plus float is incorrect? - sirocawa - Apr-16-2024
ok, thank all, i understand.
RE: python calculate float plus float is incorrect? - DeaD_EyE - Apr-16-2024
I don't know what PHP does to prevent the visible inaccuracy.
In modern programming languages, you don't see the real floating point value inclusive PHP.
An algorithm is used, to show the user a nicer representation of the floating point value.
But this does not always work (e.G. 0.2 + 0.1).
Ancient programming languages haven't used it, so the user was able to see the imprecision caused
by the conversion from 10base to binary.
The key point is, that you can't represent all decimal numbers as binary numbers.
But there are different solutions.
# slow, but precise
from decimal import Decimal
# the quotes are mandatory
# if you use float literals, it's already imprecise before Decimal is instantiated
a = Decimal("27.56")
b = Decimal("13.78")
print("Decimal:", a + b)# Scientific rounding
a = 27.56
b = 13.78
print("Round 2 decimal places:", round(a + b, 2))# floor
# ceil
from math import ceil, floor
a = 27.56
b = 13.78
print("floor:", floor(a+b))
print("ceil:", ceil(a+b))# adding the two int => int
# then divide by 100 => float
a = 2756
b = 1378
print("Int addition", (a+b) / 100)# not rounding, just formatting the output instead of changing the float itself
a = 27.56
b = 13.78
c = a + b
print(f"Output formatting: {c:.2f}")Output:Decimal: 41.34
Round 2 decimal places: 41.34
floor: 41
ceil: 42
Int addition 41.34
Output formatting: 41.34