Floating-point numbers

Numbers with a decimal point, stored in binary with about 16 significant digits - right for measuring, but slightly off for most decimal fractions, so never compared with ==.

Depth 1 · Foundation

You can use floats for measurements: / always gives a float, float(input()) reads one, round shows it to so many decimals (rounding exact halves to even), and you round only at the end.

Depths of understanding

1
Foundation
You can use floats for measurements: / always gives a float, float(input()) reads one, round shows it to so many decimals (rounding exact halves to even), and you round only at the end.
Opens after
Accumulator patternAn average is a running total divided by a count, and averages are where whole numbers stop being enough.
2
Deeper
You know why 0.1 + 0.2 is not 0.3 - floats are binary fractions with 53 digits, so most decimals are stored a tiny bit off - and so compare floats with a tolerance, keep money in whole cents, and expect small errors to add up over many steps.
Master depth 1 first