The "nan" Symbol in Python

In Python, "nan" or "NaN" is an acronym for "Not a Number". It's a special representation for an undefined or invalid numerical value.

Consider this: when you attempt to divide infinity by infinity:

x=float('infinity')
y=float('infinity')
print(x/y)

Python responds with the "nan" symbol. Why? Because the mathematical operation ∞/∞ doesn't have a defined outcome.

nan

This indicates that the result isn't a valid number It's either undefined or simply can't be represented.

Interestingly, any arithmetic involving "nan" and other numbers will invariably produce "nan

To assign this unique value to a variable, you can employ the float function:

x=float('nan')

For those seeking alternatives, the 'nan' value is accessible via the math library

import math
x = math.nan

Or, if you're working with the numpy library

import numpy as np
x = np.nan

Now, if you were to add x to 1, Python consistently returns "nan"

x+1

nan

Here's a curious aspect of 'nan': it's not considered a number in the traditional sense. In fact, it's not even equivalent to itself.

Testing its equality might surprise you:

x=float('nan')
print(x==float('nan'))

False

To effectively compare "nan", it's advisable to use the isnan() function from the math module:

import math
x=float('nan')
print(math.isnan(x))

True

For those leveraging the numpy module, a similar function exists:

import numpy as np
x=float('nan')
print(np.isnan(x))

True

To wrap things up, it's crucial to differentiate between NaN and None in Python.

While "nan" denotes an invalid or undefined numerical value, "None" represents a value's absence or a null state.