| Safe Haskell | None |
|---|---|
| Language | Haskell2010 |
Goal.Core.Util
Description
A collection of generic numerical and list manipulation functions.
Synopsis
- takeEvery :: Int -> [x] -> [x]
- breakEvery :: Int -> [x] -> [[x]]
- kFold :: Int -> [x] -> [([x], [x])]
- kFold' :: Int -> [x] -> [([x], [x], [x])]
- roundSD :: RealFloat x => Int -> x -> x
- toPi :: RealFloat x => x -> x
- circularDistance :: RealFloat x => x -> x -> x
- integrate :: Double -> (Double -> Double) -> Double -> Double -> (Double, Double)
- logistic :: Floating x => x -> x
- logit :: Floating x => x -> x
- square :: Floating x => x -> x
- triangularNumber :: Int -> Int
- average :: (Foldable f, Fractional x) => f x -> x
- weightedAverage :: (Foldable f, Fractional x) => f (x, x) -> x
- circularAverage :: (Traversable f, RealFloat x) => f x -> x
- weightedCircularAverage :: (Traversable f, RealFloat x) => f (x, x) -> x
- range :: RealFloat x => x -> x -> Int -> [x]
- discretizeFunction :: Double -> Double -> Int -> (Double -> Double) -> [(Double, Double)]
- logSumExp :: (Ord x, Floating x, Traversable f) => f x -> x
- logIntegralExp :: Traversable f => Double -> (Double -> Double) -> Double -> Double -> f Double -> Double
- traceGiven :: Show a => a -> a
- finiteInt :: KnownNat n => Finite n -> Int
- natValInt :: KnownNat n => Proxy n -> Int
- type Triangular n = Div (n * (n + 1)) 2
- data Rat (n :: Nat) (d :: Nat)
- type (/) n d = Rat n d
- ratVal :: (KnownNat n, KnownNat d) => Proxy (n / d) -> Rational
List Manipulation
breakEvery :: Int -> [x] -> [[x]] Source #
Break the list up into lists of length n.
kFold :: Int -> [x] -> [([x], [x])] Source #
Returns k (training,validation) pairs. k should be greater than or equal to 2.
kFold' :: Int -> [x] -> [([x], [x], [x])] Source #
Returns k (training,test,validation) pairs for early stopping algorithms. k should be greater than or equal to 3.
Numeric
roundSD :: RealFloat x => Int -> x -> x Source #
Rounds the number to the specified significant digit.
circularDistance :: RealFloat x => x -> x -> x Source #
Distance between two points on a circle, removing rotations.
Arguments
| :: Double | Error Tolerance |
| -> (Double -> Double) | Function |
| -> Double | Interval beginning |
| -> Double | Interval end |
| -> (Double, Double) | Integral |
Numerically integrates a 1-d function over an interval.
square :: Floating x => x -> x Source #
The square of a number (for avoiding endless default values).
triangularNumber :: Int -> Int Source #
Triangular number.
List Numerics
average :: (Foldable f, Fractional x) => f x -> x Source #
Average value of a Traversable of Fractionals.
weightedAverage :: (Foldable f, Fractional x) => f (x, x) -> x Source #
Weighted Average given a Traversable of (weight,value) pairs.
circularAverage :: (Traversable f, RealFloat x) => f x -> x Source #
Circular average value of a Traversable of radians.
weightedCircularAverage :: (Traversable f, RealFloat x) => f (x, x) -> x Source #
Weighted Circular average value of a Traversable of radians.
range :: RealFloat x => x -> x -> Int -> [x] Source #
Returns n numbers which uniformly partitions the interval [mn,mx].
discretizeFunction :: Double -> Double -> Int -> (Double -> Double) -> [(Double, Double)] Source #
Takes range information in the form of a minimum, maximum, and sample count, a function to sample, and returns a list of pairs (x,f(x)) over the specified range.
logSumExp :: (Ord x, Floating x, Traversable f) => f x -> x Source #
Given a set of values, computes the "soft maximum" by way of taking the exponential of every value, summing the results, and then taking the logarithm. Incorporates some tricks to improve numerical stability.
Arguments
| :: Traversable f | |
| => Double | Error Tolerance |
| -> (Double -> Double) | Function |
| -> Double | Interval beginning |
| -> Double | Interval end |
| -> f Double | Samples (for approximating the max) |
| -> Double | Log-Integral-Exp |
Given a function, computes the "soft maximum" of the function by computing the integral of the exponential of the function, and taking the logarithm of the result. The maximum is first approximated on a given set of samples to improve numerical stability. Pro tip: If you want to compute the normalizer of a an exponential family probability density, provide the unnormalized log-density to this function.
Tracing
traceGiven :: Show a => a -> a Source #
Runs traceShow on the given element.