invcot (example 3.9)

Average Accuracy: 6.4% → 99.5%

Time: 14.2s

Precision: binary64

Cost: 13632

\[-0.026 < x \land x < 0.026\]

Math

\[\frac{1}{x} - \frac{1}{\tan x} \]

↓

\[0.3333333333333333 \cdot x + \left(0.0021164021164021165 \cdot {x}^{5} + 0.022222222222222223 \cdot {x}^{3}\right) \]

FPCore

(FPCore (x) :precision binary64 (- (/ 1.0 x) (/ 1.0 (tan x))))

↓

(FPCore (x)
 :precision binary64
 (+
  (* 0.3333333333333333 x)
  (+
   (* 0.0021164021164021165 (pow x 5.0))
   (* 0.022222222222222223 (pow x 3.0)))))

C

double code(double x) {
	return (1.0 / x) - (1.0 / tan(x));
}

↓

double code(double x) {
	return (0.3333333333333333 * x) + ((0.0021164021164021165 * pow(x, 5.0)) + (0.022222222222222223 * pow(x, 3.0)));
}

Fortran

real(8) function code(x)
    real(8), intent (in) :: x
    code = (1.0d0 / x) - (1.0d0 / tan(x))
end function

↓

real(8) function code(x)
    real(8), intent (in) :: x
    code = (0.3333333333333333d0 * x) + ((0.0021164021164021165d0 * (x ** 5.0d0)) + (0.022222222222222223d0 * (x ** 3.0d0)))
end function

Java

public static double code(double x) {
	return (1.0 / x) - (1.0 / Math.tan(x));
}

↓

public static double code(double x) {
	return (0.3333333333333333 * x) + ((0.0021164021164021165 * Math.pow(x, 5.0)) + (0.022222222222222223 * Math.pow(x, 3.0)));
}

Python

def code(x):
	return (1.0 / x) - (1.0 / math.tan(x))

↓

def code(x):
	return (0.3333333333333333 * x) + ((0.0021164021164021165 * math.pow(x, 5.0)) + (0.022222222222222223 * math.pow(x, 3.0)))

Julia

function code(x)
	return Float64(Float64(1.0 / x) - Float64(1.0 / tan(x)))
end

↓

function code(x)
	return Float64(Float64(0.3333333333333333 * x) + Float64(Float64(0.0021164021164021165 * (x ^ 5.0)) + Float64(0.022222222222222223 * (x ^ 3.0))))
end

MATLAB

function tmp = code(x)
	tmp = (1.0 / x) - (1.0 / tan(x));
end

↓

function tmp = code(x)
	tmp = (0.3333333333333333 * x) + ((0.0021164021164021165 * (x ^ 5.0)) + (0.022222222222222223 * (x ^ 3.0)));
end

Wolfram

code[x_] := N[(N[(1.0 / x), $MachinePrecision] - N[(1.0 / N[Tan[x], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]

↓

code[x_] := N[(N[(0.3333333333333333 * x), $MachinePrecision] + N[(N[(0.0021164021164021165 * N[Power[x, 5.0], $MachinePrecision]), $MachinePrecision] + N[(0.022222222222222223 * N[Power[x, 3.0], $MachinePrecision]), $MachinePrecision]), $MachinePrecision]), $MachinePrecision]

Target

Original	6.4%
Target	99.9%
Herbie	99.5%

\[\begin{array}{l} \mathbf{if}\;\left|x\right| < 0.026:\\ \;\;\;\;\frac{x}{3} \cdot \left(1 + \frac{x \cdot x}{15}\right)\\ \mathbf{else}:\\ \;\;\;\;\frac{1}{x} - \frac{1}{\tan x}\\ \end{array} \]

Derivation

1. Initial program 6.4%

   \[\frac{1}{x} - \frac{1}{\tan x} \]

2. Taylor expanded in x around 0 99.5%

   \[\leadsto \color{blue}{0.3333333333333333 \cdot x + \left(0.0021164021164021165 \cdot {x}^{5} + 0.022222222222222223 \cdot {x}^{3}\right)} \]

3. Final simplification 99.5%

   \[\leadsto 0.3333333333333333 \cdot x + \left(0.0021164021164021165 \cdot {x}^{5} + 0.022222222222222223 \cdot {x}^{3}\right) \]

Alternatives

Alternative 1
Accuracy	99.4%
Cost	6912

\[0.3333333333333333 \cdot x + 0.022222222222222223 \cdot {x}^{3} \]

Alternative 2
Accuracy	98.9%
Cost	192

\[0.3333333333333333 \cdot x \]

Reproduce

herbie shell --seed 2023126 
(FPCore (x)
  :name "invcot (example 3.9)"
  :precision binary64
  :pre (and (< -0.026 x) (< x 0.026))

  :herbie-target
  (if (< (fabs x) 0.026) (* (/ x 3.0) (+ 1.0 (/ (* x x) 15.0))) (- (/ 1.0 x) (/ 1.0 (tan x))))

  (- (/ 1.0 x) (/ 1.0 (tan x))))