Skip to content

Agilearn

math module reference

accuser/agilearn

`math` module reference¶

The everyday math functions on one page. All operate on float and return float unless noted; the integer functions (isqrt, comb, perm, factorial, gcd, lcm) stay exact in int.

import math

Powers, roots, and logarithms¶

Function	Returns	Example
`sqrt(x)`	square root (float)	`sqrt(16)` → `4.0`
`isqrt(n)`	integer square root, floored	`isqrt(17)` → `4`
`cbrt(x)`	cube root (3.11+)	`cbrt(27)` → `3.0`
`pow(x, y)`	`x ** y` as float	`pow(2, 10)` → `1024.0`
`exp(x)`	e^x	`exp(1)` → `2.718…`
`log(x[, base])`	natural log, or log to `base`	`log(8, 2)` → `3.0`
`log2(x)` / `log10(x)`	log base 2 / base 10	`log10(1000)` → `3.0`
`hypot(*coords)`	Euclidean norm √(Σx²)	`hypot(3, 4)` → `5.0`

Use the built-in **/pow for integer powers (exact, any size); use math.pow only when you want a float result.

Rounding (return `int`)¶

Function	Rounds	Example
`floor(x)`	toward −∞	`floor(-3.2)` → `-4`
`ceil(x)`	toward +∞	`ceil(3.2)` → `4`
`trunc(x)`	toward zero	`trunc(-3.7)` → `-3`

The built-in round is separate — it rounds half to even and returns the same type. See round numbers correctly.

Combinatorics and integer maths (return `int`)¶

Function	Returns	Example
`factorial(n)`	n!	`factorial(5)` → `120`
`comb(n, k)`	combinations, n choose k	`comb(5, 2)` → `10`
`perm(n, k)`	permutations	`perm(5, 2)` → `20`
`gcd(*ints)`	greatest common divisor	`gcd(12, 18)` → `6`
`lcm(*ints)`	lowest common multiple (3.9+)	`lcm(4, 6)` → `12`
`prod(iterable, start=1)`	product of items	`prod([1,2,3,4])` → `24`

Trigonometry¶

Angles are in radians. Convert with degrees/radians.

Function	Notes
`sin(x)`, `cos(x)`, `tan(x)`	basic trig
`asin`, `acos`, `atan`, `atan2(y, x)`	inverses; `atan2` knows the quadrant
`degrees(x)` / `radians(x)`	convert between degrees and radians
`sinh`, `cosh`, `tanh`	hyperbolic

Constants¶

Name	Value
`math.pi`	`3.141592653589793`
`math.e`	`2.718281828459045`
`math.tau`	`6.283185307179586` (2π)
`math.inf`	floating-point infinity
`math.nan`	floating-point not-a-number

Testing and comparing¶

Function	Returns
`isnan(x)`	`True` if `x` is `nan` (the only reliable test)
`isinf(x)`	`True` if `x` is `±inf`
`isfinite(x)`	`True` if neither `nan` nor `inf`
`isclose(a, b, *, rel_tol=1e-9, abs_tol=0.0)`	safe float equality
`copysign(x, y)`	`x` with the sign of `y`
`fabs(x)`	absolute value as float
`fsum(iterable)`	accurate sum of floats (tracks error)
`fmod(x, y)`	C-style remainder (sign of `x`; differs from `%`)

For arbitrary precision (beyond float), do the maths in Decimal — it has its own sqrt, ln, log10, and exp methods.