packages feed

ad 3.2.1 → 3.2.2

raw patch · 3 files changed

+13/−1 lines, 3 filesPVP ok

version bump matches the API change (PVP)

API changes (from Hackage documentation)

+ Numeric.AD: conjugateGradientAscent :: (Traversable f, Fractional a, Ord a) => (forall s. Mode s => f (AD s a) -> AD s a) -> f a -> [f a]
+ Numeric.AD: conjugateGradientDescent :: (Traversable f, Fractional a, Ord a) => (forall s. Mode s => f (AD s a) -> AD s a) -> f a -> [f a]
+ Numeric.AD: gradientAscent :: (Traversable f, Fractional a, Ord a) => (forall s. Mode s => f (AD s a) -> AD s a) -> f a -> [f a]
+ Numeric.AD: gradientDescent :: (Traversable f, Fractional a, Ord a) => (forall s. Mode s => f (AD s a) -> AD s a) -> f a -> [f a]

Files

CHANGELOG.markdown view
@@ -1,3 +1,7 @@+3.2.2+-----+* Export of the `conjugateGradientDescent` and `gradientDescent` from `Numeric.AD`+ 3.2.1 --- * `conjugateGradientDescent` now stops before it starts returning NaN results.
ad.cabal view
@@ -1,5 +1,5 @@ name:         ad-version:      3.2.1+version:      3.2.2 license:      BSD3 license-File: LICENSE copyright:    (c) Edward Kmett 2010-2012,
src/Numeric/AD.hs view
@@ -108,6 +108,12 @@     -- * Maclaurin Series (Tower)     , maclaurin     , maclaurin0++    -- * Gradient Descent+    , gradientDescent+    , gradientAscent+    , conjugateGradientDescent+    , conjugateGradientAscent     ) where  import Data.Traversable (Traversable)@@ -136,6 +142,8 @@ import qualified Numeric.AD.Mode.Sparse as Sparse import Numeric.AD.Mode.Sparse     ( grads, jacobians, hessian', hessianF')++import Numeric.AD.Newton  -- | Calculate the Jacobian of a non-scalar-to-non-scalar function, automatically choosing between forward and reverse mode AD based on the number of inputs and outputs. --