33. Dealing with NaN

By Bernd Klein. Last modified: 03 Feb 2025.

Introduction

NaN was introduced, at least officially, by the IEEE Standard for Floating-Point Arithmetic (IEEE 754). It is a technical standard for floating-point computation established in 1985 - many years before Python was invented, and even a longer time befor Pandas was created - by the Institute of Electrical and Electronics Engineers (IEEE). It was introduced to solve problems found in many floating point implementations that made them difficult to use reliably and portably.

This standard added NaN to the arithmetic formats: "arithmetic formats: sets of binary and decimal floating-point data, which consist of finite numbers (including signed zeros and subnormal numbers), infinities, and special 'not a number' values (NaNs)"