name: hasty-hamiltonian
version: 1.3.2
synopsis: Speedy traversal through parameter space.
homepage: http://github.com/jtobin/hasty-hamiltonian
license: MIT
license-file: LICENSE
author: Jared Tobin
maintainer: jared@jtobin.ca
category: Numeric
build-type: Simple
cabal-version: >= 1.10
Description:
Gradient-based traversal through parameter space.
.
This implementation of HMC algorithm uses 'lens' as a means to operate over
generic indexed traversable functors, so you can expect it to work if your
target function takes a list, vector, map, sequence, etc. as its argument.
.
If you don't want to calculate your gradients by hand you can use the
handy <https://hackage.haskell.org/package/ad ad> library for automatic
differentiation.
.
Exports a 'mcmc' function that prints a trace to stdout, a 'chain' function
for collecting results in memory, and a 'hamiltonian' transition operator
that can be used more generally.
.
> import Numeric.AD (grad)
> import Numeric.MCMC.Hamiltonian
>
> target :: RealFloat a => [a] -> a
> target [x0, x1] = negate ((x0 + 2 * x1 - 7) ^ 2 + (2 * x0 + x1 - 5) ^ 2)
>
> gTarget :: [Double] -> [Double]
> gTarget = grad target
>
> booth :: Target [Double]
> booth = Target target (Just gTarget)
>
> main :: IO ()
> main = withSystemRandom . asGenIO $ mcmc 10000 0.05 20 [0, 0] booth
Source-repository head
Type: git
Location: http://github.com/jtobin/hasty-hamiltonian.git
library
default-language: Haskell2010
ghc-options:
-Wall
exposed-modules:
Numeric.MCMC.Hamiltonian
build-depends:
base >= 4 && < 6
, kan-extensions >= 5 && < 6
, mcmc-types >= 1.0.1
, mwc-probability >= 2.0 && < 3
, lens >= 4 && < 5
, pipes >= 4 && < 5
, primitive >= 0.5 && < 1.0
, transformers >= 0.5 && < 1.0
Test-suite booth
type: exitcode-stdio-1.0
hs-source-dirs: test
main-is: Booth.hs
default-language: Haskell2010
ghc-options:
-rtsopts
build-depends:
ad >= 4 && < 5
, base >= 4 && < 6
, mwc-probability >= 2.0 && < 3
, hasty-hamiltonian