Average Error: 1.7 → 1.5
Time: 1.6m
Precision: 64
Internal Precision: 576
\[\frac{x \cdot e^{\left(y \cdot \log z + \left(t - 1.0\right) \cdot \log a\right) - b}}{y}\]
↓
\[\begin{array}{l} \mathbf{if}\;y \le -651945089936.9535 \lor \neg \left(y \le 1.511346421991279 \cdot 10^{-32}\right):\\ \;\;\;\;\frac{x}{\frac{y}{e^{\left(\log a \cdot \left(t - 1.0\right) + \log z \cdot y\right) - b}}}\\ \mathbf{else}:\\ \;\;\;\;\frac{\frac{x}{\frac{1}{\frac{{z}^{y} \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}{y}}}}{{a}^{1.0}}\\ \end{array}\]

Error

Bits error versus x

Bits error versus y

Bits error versus z

Bits error versus t

Bits error versus a

Bits error versus b

Derivation

  1. Split input into 2 regimes

  2. if y < -651945089936.9535 or 1.511346421991279e-32 < y

    1. Initial program 0.1

      \[\frac{x \cdot e^{\left(y \cdot \log z + \left(t - 1.0\right) \cdot \log a\right) - b}}{y}\]
    2. Using strategy rm

    3. Applied associate-/l*0.1

      \[\leadsto \color{blue}{\frac{x}{\frac{y}{e^{\left(y \cdot \log z + \left(t - 1.0\right) \cdot \log a\right) - b}}}}\]

    if -651945089936.9535 < y < 1.511346421991279e-32

    1. Initial program 3.3

      \[\frac{x \cdot e^{\left(y \cdot \log z + \left(t - 1.0\right) \cdot \log a\right) - b}}{y}\]
    2. Using strategy rm

    3. Applied associate-/l*3.4

      \[\leadsto \color{blue}{\frac{x}{\frac{y}{e^{\left(y \cdot \log z + \left(t - 1.0\right) \cdot \log a\right) - b}}}}\]

    4. Taylor expanded around inf 3.4

      \[\leadsto \frac{x}{\frac{y}{\color{blue}{e^{1.0 \cdot \log \left(\frac{1}{a}\right) - \left(\log \left(\frac{1}{z}\right) \cdot y + \left(b + t \cdot \log \left(\frac{1}{a}\right)\right)\right)}}}}\]

    5. Simplified2.5

      \[\leadsto \frac{x}{\frac{y}{\color{blue}{\left({a}^{\left(-1.0\right)} \cdot {z}^{y}\right) \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}}}\]
    6. Using strategy rm

    7. Applied pow-neg2.5

      \[\leadsto \frac{x}{\frac{y}{\left(\color{blue}{\frac{1}{{a}^{1.0}}} \cdot {z}^{y}\right) \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}}\]

    8. Applied associate-*l/2.5

      \[\leadsto \frac{x}{\frac{y}{\color{blue}{\frac{1 \cdot {z}^{y}}{{a}^{1.0}}} \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}}\]

    9. Applied associate-*l/2.5

      \[\leadsto \frac{x}{\frac{y}{\color{blue}{\frac{\left(1 \cdot {z}^{y}\right) \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}{{a}^{1.0}}}}}\]

    10. Applied associate-/r/2.5

      \[\leadsto \frac{x}{\color{blue}{\frac{y}{\left(1 \cdot {z}^{y}\right) \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}} \cdot {a}^{1.0}}}\]

    11. Applied associate-/r*2.9

      \[\leadsto \color{blue}{\frac{\frac{x}{\frac{y}{\left(1 \cdot {z}^{y}\right) \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}}}{{a}^{1.0}}}\]
    12. Using strategy rm

    13. Applied clear-num2.9

      \[\leadsto \frac{\frac{x}{\color{blue}{\frac{1}{\frac{\left(1 \cdot {z}^{y}\right) \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}{y}}}}}{{a}^{1.0}}\]
  3. Recombined 2 regimes into one program.

  4. Final simplification1.5

    \[\leadsto \begin{array}{l} \mathbf{if}\;y \le -651945089936.9535 \lor \neg \left(y \le 1.511346421991279 \cdot 10^{-32}\right):\\ \;\;\;\;\frac{x}{\frac{y}{e^{\left(\log a \cdot \left(t - 1.0\right) + \log z \cdot y\right) - b}}}\\ \mathbf{else}:\\ \;\;\;\;\frac{\frac{x}{\frac{1}{\frac{{z}^{y} \cdot e^{(t \cdot \left(\log a\right) + \left(-b\right))_*}}{y}}}}{{a}^{1.0}}\\ \end{array}\]

Runtime

Time bar (total: 1.6m)Debug logProfile

BaselineHerbieOracleSpan%
Regimes1.81.50.01.815.5%
herbie shell --seed 2018263 +o rules:numerics
(FPCore (x y z t a b)
  :name "Numeric.SpecFunctions:incompleteBetaWorker from math-functions-0.1.5.2"
  (/ (* x (exp (- (+ (* y (log z)) (* (- t 1.0) (log a))) b))) y))