README.md (1632B)
1 2 # measurable 3 4 [](https://github.com/jtobin/measurable/blob/master/LICENSE) 5 6 **measurable** is a simple shallowly-embedded DSL for dealing with measures. 7 8 It uses a `Measure` synonym for a standard continuation type with a restricted 9 output type and no `callCC` implementation. You can construct measures from 10 samples, mass/density functions, or even sampling functions. 11 12 Construct image measures by `fmap`-ing measurable functions over them, or 13 create new measures from existing ones by measure convolution and friends 14 provided by a simple `Num` instance enabled by an `Applicative` instance. 15 Create measures from graphs of other measures using the `Monad` instance and 16 do-notation. 17 18 Query measures by integrating measurable functions against them. Extract 19 moments, cumulative density functions, or probabilities. 20 21 You can check out a few blog posts I wrote about the theoretical foundations 22 and implementation of the library here: 23 24 * [Foundations of the Giry Monad][foun] 25 * [Implementing the Giry Monad][impl] 26 * [The Applicative Structure of the Giry Monad][appl] 27 28 A more polished and extended version of the above appears in chapter three of 29 [my dissertation][diss]. 30 31 Caveat: while fun to play with, and rewarding to see how measures fit together, 32 measure operations as nested integrals are exponentially complex. Don't expect 33 them to scale very far! 34 35 [diss]: https://jtobin.io/assets/jtobin-dissertation.pdf 36 [foun]: https://jtobin.io/giry-monad-foundations 37 [impl]: https://jtobin.io/giry-monad-implementation 38 [appl]: https://jtobin.io/giry-monad-applicative