úÎ9ã4îK      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJNonePositive number. 1Interpretations chosen in the given context with ! corresponding positive weights. A sentence of s. 'A word parametrized over the tag type. Orthographic form. Set of word interpretations. +Retrieve the most probable interpretation. :A word is considered to be known when the set of possible  interpretations is not empty. K    K None A token. *Interpretations with disambiguation info. No space, space or newline. *Extract information relevant for tagging. ?Mark all interpretations with tag component beeing a member of / the given choice set with disamb annotations. 6Add new interpretations with given disamb annotation. Add new interpretations with None# base and given disamb annotation. L Printing. An infix synonym for M. # !"NOPQL#$%&RSTUV !"#$%& !"#$%& !"NOPQL#$%&RSTUVNone '7A guesser represented by the conditional random field. )The CRF model * The tag indicating unkown words +@An observation consist of an index (of list type) and an actual  observation value. ,?A schema is a block of the Ox computation performed within the = context of the sentence and the absolute sentence position. -CThe Ox monad specialized to word token type and text observations. E TODO: Move to monad-ox package from here and from the nerf library. /0Schematize the input sentence with according to . rules. 0Determine the k, most probable labels for each unknown word  in the sentence. 1Tag the file. 2%TODO: Abstract over the format type. W&Schematized data from the plain file. '()*+,-./01Guesser argument Guesser itself File to tag (plain format) 2SGD parameters !The tag indicating unknown words Train file (plain format) Maybe eval file WX '()*+,-./012 -,+./'()*012 '()*+,-./012WXNone3The disambiguation model. 5@An observation consist of an index (of list type) and an actual  observation value. 6?A schema is a block of the Ox computation performed within the = context of the sentence and the absolute sentence position. 7CThe Ox monad specialized to word token type and text observations. E TODO: Move to monad-ox package from here and from the nerf library. 8A tag with optional POS. <A tier description. >$Does it include the part of speech? ?Tier grammatical attributes. @Select tier attributes. B+Schematize the input sentence according to A rules. DSplit tags between two layers. ( TODO: Add support for multiple layers. F,Determine the most probable label sequence. HAUnsplit the list of tag pairs. TODO: It can be done without the  help of original word. ITag the file. J%TODO: Abstract over the format type. Y&Schematized data from the plain file. 3Z[\]456789:;<=>?@ABCDEFGHITag indicating unknown words File to tag (plain format) JSGD parameters 'File with positional tagset definition !The tag indicating unknown words Tiered tagging configuration Train file (plain format) Maybe eval file Y^_`3456789:;<=>?@ABCDEFGHIJ<=>?89:;@DE765AB4CHG3FIJ3Z[\]456789:;<=>?@ABCDEFGHIJY^_`a        !"#$%&&'()*+,-./012)*+334566789,-:;<=>?/0@ABCDEFGHIJKLML1'NOPQRSconcraft-0.2.0NLP.Concraft.MorphosyntaxNLP.Concraft.PlainNLP.Concraft.GuessNLP.Concraft.DisambPositive unPositiveChoiceSentWordorthtagsmapWord<+> mapChoice mkPositivebestknownInterpbasetagTokenspaceinterpsSpaceNewLineNonefromTokchoose addInterpsaddNones readPlain parsePlain parseSent writePlain showPlainshowSentshowWordGuessercrfignObSchemaOxschema schematizeguesstagFilelearnDisambTierConfTagposattsTierwithPoswithAttsselecttear splitWord splitSentdisambdeTearsdeTear<> Data.Monoidmappend parseWord parseInterp parseHeader parseSpace buildSent buildWord buildInterps buildSpace buildKnownschemed$fBinaryGuessertagsettierConf$fBinaryDisamb $fBinaryTag $fBinaryTier