NLP.Probability.Observation

This module provides a simple way to collect observations (counts), particularly within a monoid. Use observation for each observed event and mappend for combining observations. Finally finish before estimating probabilities.

type Count = Double

Source

data Counts event

Source

Observations over a set of events. The param event must be an instance of class Event

Instances

(Event event, Show event) => Show (Counts event)

Event event => Monoid (Counts event)

(Event event, Binary event, Binary (EventMap event event Count)) => Binary (Counts event)

(Event event, NFData event) => NFData (Counts event)

(Event event, Pretty event) => Pretty (Counts event)

class Map (EventMap event) event => Event event

Source

Trivial type family for events. Just use EventMap = M.Map for most cases. Allows clients to specify the type of map used, when efficiency is important.

Associated Types

type EventMap event :: * -> * -> *

Source

Methods

Instances

Event Int

Event String

Event Word

observation :: Event event => event -> Counts event

Source

Observation of a single event

observations :: Map (EventMap event) event => event -> Count -> Counts event

Source

inc :: Event e => Counts e -> e -> Count -> Counts e

Source

Manually increment the count of an event

data Observed event

Source

Constructors

Observed

observed :: EventMap event event Count
total :: Double	Gives the total number of observations sum_a C(a)
unique :: Count	Gives the total number of events observed at least once {a \| C(a) > 1}

finish :: Event event => Counts event -> Observed event

Source

Finish a set of offline observations so that they can be used to estimate likelihood

showObsPretty :: (Event event, Monad m) => (event -> m Doc) -> Counts event -> m Doc

Source

Produced by Haddock version 2.6.0