Average Error: 0.1 → 0.1
Time: 1.4s
Precision: binary64
\[1 - \frac{1}{1 + e^{A \cdot f + B}}\]
↓
\[1 - \frac{1}{1 + e^{A \cdot f + B}}\]
1 - \frac{1}{1 + e^{A \cdot f + B}}
↓
1 - \frac{1}{1 + e^{A \cdot f + B}}
double code(double A, double f, double B) {
	return ((double) (1.0 - ((double) (1.0 / ((double) (1.0 + ((double) exp(((double) (((double) (A * f)) + B))))))))));
}
↓
double code(double A, double f, double B) {
	return ((double) (1.0 - ((double) (1.0 / ((double) (1.0 + ((double) exp(((double) (((double) (A * f)) + B))))))))));
}

Error

Bits error versus A

Bits error versus f

Bits error versus B

Try it out

Your Program's Arguments

Results

Enter valid numbers for all inputs

Derivation

  1. Initial program 0.1

    \[1 - \frac{1}{1 + e^{A \cdot f + B}}\]

  2. Final simplification0.1

    \[\leadsto 1 - \frac{1}{1 + e^{A \cdot f + B}}\]

Reproduce

herbie shell --seed 2020153 
(FPCore (A f B)
  :name "(- 1 (/ 1 (+ 1 (exp (+ (* A f) B)))))"
  :precision binary64
  (- 1.0 (/ 1.0 (+ 1.0 (exp (+ (* A f) B))))))