Package Info

ghc-hasty-hamiltonian


Speedy traversal through parameter space


Development/Libraries/Haskell

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.

License: MIT
URL: https://hackage.haskell.org/package/hasty-hamiltonian

Categories

Releases

Package Version Update ID Released Package Hub Version Platforms Subpackages
1.3.0-bp150.2.3 info GA Release 2018-08-01 15
  • AArch64
  • ghc-hasty-hamiltonian
  • ghc-hasty-hamiltonian-devel
1.3.0-bp150.2.5 info GA Release 2018-07-30 15
  • x86-64
  • ghc-hasty-hamiltonian
  • ghc-hasty-hamiltonian-devel
1.3.0-bp150.2.6 info GA Release 2018-07-31 15
  • ppc64le
  • ghc-hasty-hamiltonian
  • ghc-hasty-hamiltonian-devel