Avg

avg calculates the average of the numbers in an array.

Overview

Use avg to find the average value in a collection of numbers.

temperatures | [68, 72, 75, 69]
average | avg(temperatures)

:say(average)  /// 71

temperatures | [68, 72, 75, 69]
average | avg(temperatures)

:say(average)  /// 71

Goblin adds the values together and divides the total by the number of values.

Number Types

avg accepts an array containing numeric values:

avg([8, 10, 12])       /// 10
avg([1.5, 2.5, 5.0])  /// 3

avg([8, 10, 12])       /// 10
avg([1.5, 2.5, 5.0])  /// 3

Integers, floats, percentages, and big numbers are numeric values.

For an array of integers, floats, or percentages, avg returns a float.

If the array contains a big number, avg uses big-number arithmetic and returns a big number.

Empty Arrays

The average of an empty array is 0.0:

avg([])  /// 0.0

avg([])  /// 0.0

This allows an empty collection to be averaged without raising an error.

Mixed Numbers

Integers, floats, and percentages may be used together:

measurements | [10, 12.5, 15]

avg(measurements)  /// 12.5

measurements | [10, 12.5, 15]

avg(measurements)  /// 12.5

Big-number calculations have stricter conversion rules. In the current interpreter, an array containing a big number cannot also contain an integer. Convert the integer to a big number first:

values | [big(10), big(20), big(30)]

avg(values)

values | [big(10), big(20), big(30)]

avg(values)

Signature

avg(numbers)

avg(numbers)

avg returns a float for ordinary numeric arrays and a big number when big-number arithmetic is used.

Errors

avg requires exactly one argument:

avg([10, 20], [30, 40])
/// error: R0301 wrong-arity

avg([10, 20], [30, 40])
/// error: R0301 wrong-arity

The argument must be an array:

avg(10)
/// error: R0402 array-expected

avg(10)
/// error: R0402 array-expected

Every value in the array must be a compatible number:

avg([10, "twelve", 14])
/// error: R0200 numeric-expected

avg([10, "twelve", 14])
/// error: R0200 numeric-expected

A non-finite float cannot be converted for a big-number calculation:

/// error: T0320 non-finite-float

/// error: T0320 non-finite-float