Floats

Go has two floating-point types, float32 and float64. Go uses the IEEE 754 specification, giving a large range and limited precision. Unless you have to be compatible with an existing format, use float64 since float literals defaul value is float64.

Precision Problems

Similar to many programming languages, Go's floating-point numbers have an extensive range, yet they cannot precisely represent every value within that spectrum; rather, they store the closest approximation. Given their inherent imprecision, floating-point numbers are suitable only for scenarios where approximate values suffice or where the nuances of floating-point arithmetic are well understood.

Although Go permits the use of == and != operators for comparing floats, it's advisable to avoid doing so. Due to the inherent imprecision of floats, two floating-point values may not be considered equal even when they appear identical. Instead, consider defining a maximum permissible variance and comparing whether the difference between two floats falls below this threshold. This threshold value (often referred to as epsilon), should be chosen based on the precision requirements of your application.

Help to define variance value

Checkout: Floating-Point Guide