úÎ1€,=@      !"#$%&'()*+,-./0123456789:;<=>?  Safe-Inferred' chunk n xs' splits xs into n chunks None Safe-Inferred@@None'UniqueKey val key'7 is a monad for a calculation of a mapping unique keys  key onto values val /Get map of unique keys to values !Get map of values to unique keys AFind the unique key for value val or B if the value is unknown Run a 2, returning the result and the associated key map Run a 2, returning the result and the associated key map CDA    CDA NoneEE NoneFFNone A Dirichlet prior G!Make error handling a bit easier Construct an asymmetric Alpha H;Compute the normalizer of the likelihood involving alphas,  (product_k gamma(alpha_k)) / gamma(sum_k alpha_k) ' alphaDomain a' is the domain of prior a 'alphaOf alpha k' is the value of element k in prior alpha 'sumAlpha alpha' is the sum of all alphas Set a particular alpha element 'alphaToMeanPrecision a' is the mean/&precision representation of the prior a 'meanPrecisionToAlpha m p' is a prior with mean m and precision I 1Symmetrize a Dirichlet prior (such that mean=0) JCTurn a symmetric alpha into an asymmetric alpha. For internal use. Pretty-print a Dirichlet prior  KLMNOPQGHJR   NKOPQLMQGHJRNone  !"STUV#  !"#  !"# !"STUV#None$CA distribution for which a full conditional factor can be produced $%&'()*+$%&'()*++'()*$%&$%&'()*+ None !"#$%&'()*+None ,' Multinom a'1 represents multinomial distribution over domain a. ; Optionally, this can include a collapsed Dirichlet prior.  'Multinom alpha count total'' is a multinomial with Dirichlet prior  with symmetric parameter alpha, ... W!Make error handling a bit easier X'symMultinomFromPrecision d p' is a symmetric Dirichlet/multinomial over a  domain d with precision p Y'dirMultiFromMeanPrecision m p' is an asymmetric Dirichlet/ multinomial  over a domain d with mean m and precision p 6Create a symmetric Dirichlet/ multinomial 7A multinomial without a prior 8Create an asymmetric Dirichlet/"multinomial from items and alphas ZCreate a Dirichlet/multinomial with a given prior :0Probabilities sorted decreasingly <Update the prior of a Dirichlet/ multinomial [5Relative tolerance in precision for prior estimation \0Estimate the prior alpha from a set of Dirichlet/ multinomials $,]^_-`./012Wab345XY678Zc9:;<[=>\?def,-./0123456789:;<=>?,-././867.-/502134;<?=>9:, _]-`./^`./021Wab345XY678Zc9:;<[=>\?defg   !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQ R STUVWXYZ[\]^_`abcdTefghi8jklmnopqrsbayes-stack-0.2.0.1Data.Sequence.ChunkData.Random.SequenceBayesStack.UniqueKeyBayesStack.DirichletBayesStack.Core.GibbsBayesStack.Core.TypesBayesStack.DirMultiBayesStack.TupleEnumData.Serialize.LogFloatData.Serialize.EnumMapBayesStack.CorechunkrandomElementT UniqueKeyT UniqueKey getKeyMap getValueMap getUniqueKey runUniqueKey runUniqueKeyTrunUniqueKeyT' runUniqueKey'mapTraversable DirPrecisionDirMeanAlphasymAlpha asymAlpha setSymAlpha alphaDomainalphaNormalizeralphaOfsumAlpha setAlphaOfalphaToMeanPrecisionmeanPrecisionToAlphasymmetrizeAlpha prettyAlphaWrappedUpdateUnit WrappedUU UpdateUnit ModelStateSetting fetchSetting evolveSetting updateSetting gibbsUpdateFullConditionable FCContext sampleProb HasLikelihoodLContext likelihoodprob ProbabilityMultinomdmAlphadmTotaldmDomainSetUnsetUnsetSet decMultinom incMultinom setMultinom symDirMultimultinomdirMulti probabilitiesdecProbabilitiesprettyMultinom updatePriorreestimatePriorsreestimateSymPriors estimatePrior $fEnum(,) findUniqueKeybase Data.MaybeNothing popUniqueKey$fSerializeLogFloat$fSerializeEnumMapcheckNaN alphaNorm gamma-0.9.0.2 Math.GammapasymmetrizeAlphaaAlphas aSumAlphasSymAlphaaDomainaAlphaaNorm$fSerializeAlpha updateUnit updateWorker diffWorker labelMyThreadsymDirMultiFromPrecisiondirMultiFromPrecisiondirMultiFromAlpha estimationTolestimatePrior'dmProbsDirMultidmCountsmaybeIncmaybeDec dmGetCounts$fFullConditionableMultinom$fHasLikelihoodMultinom$fSerializeMultinom