Understanding Python Numeric Types

Internally, Integers are represented using base-2 (binary) digits.
It can store signed and unsigned numbers
Here int object uses a variable number of bits, hence it can use 4 bytes, 8 bytes, 12 bytes, etc. But it is limited to the amount of memory available

Float data is represented in IEEE 754 double-precision binary float format. It uses fixed number of bytes and hence are rational numbers. They can be represented as base-10 integers and fractions
Float can be compared to be equal (i.e equality test) using a math library's isclose function. This function gives us the flexibility to compare based on a tolerance value. It can either be compared using abs_tol i.e. absolute tolerance or using rel_tol i.e. relative tolerance
abs_tol is a tolerance value calculated using absolute difference of two floats. For example, if a & b are two floats then: tol = abs(a) - abs(b) . So if absolute difference is less than tol, then those two floats are close and hence is_close returns True
rel_tol is a tolerance value calcuated using rel_tol percentage of max of two floats (absolute). For example, if a & b are two floats then: tol = rel_tol/100 * max(abs(a), abs(b)). So if absolute difference is less than tol, then those two floats are close and hence is_close returns True

Python provides in-built functions like bin(), hex(), oct() to convert decimal numbers to binary, hexadecimal, octal formats respectively
A binary number is prefixed with 0b, hexadecimal number is prefixed with 0x and octal number with 0o

Python provides various math functions like round() , floor(), ceil(), trunc()
round() function rounds off a decimal to a the closest integer. It also takes precision as an argument to control the decimal values. In case of ties, we would expect numbers to round away from zero. But this doesn't happen. Python doesn bankers rounding i.e. rounds to the nearest value, with ties rounded to the nearest value with an even least significant digit
trunc() function truncates a floating point number to just the integer part. For example trunc(1.5) will be perform truncation on1.5 to 1
floor() function converts a floating point number to closest integer away from zero (But this is not always the case in python!)
ceil() function is opposite of floor, where floating point number is converted to closest integer closer to zero

Because of intricacies of representing floating point number in IEEE format, they are no very accurately represented
Hence python has Fractions and Decimals libraries where floating point can be very closely represented in accurate format