y|i      !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefgh None9An atomic part of morphosyntactic tag with optional POS. A tier description. $Does it include the part of speech? Tier grammatical attributes. iSelect tier attributes. jSplit the positional tag. ijkl ijijklNone CRF model data. mCRF feature map. n9Transition map restricted to a particular tagging layer. o5Codec internal data. The first component is used to : encode observations of type a, the second one is used to  encode labels of type [b]. p Feature. q Sublabel. r Observation. s"Feature generation for complex [Lb] label type. t)Codec dependes on the number of layers. uDummy feature index. v0Codec specification given the number of layers. w<Train the CRF using the stochastic gradient descent method. @ Use the provided feature selection function to determine model  features. x'Find the most probable label sequence. *yz{|m}~nopqrstuvwNumber of tagging layers  Args for SGD Training data  action Maybe evalation data Resulting model xyz{|pqrwxyz{|m}~nopqrstuvwxNone 5A set with a non-negative weight assigned to each of  its elements. 2A sentence with original, textual representation.  A sentence. Orthographic form. Out-of-vocabulary (OOV) word. 8A segment parametrized over a word type and a tag type. 9A word represented by the segment. Typically it will be  an instance of the  class. 2A set of interpretations. To each interpretation 0 a weight of appropriateness within the context  is assigned.  Map function over segment tags.  Interpretations of the segment.  Interpretations of the segment. !Map function over sentence tags. !Map function over sentence tags. @Make a weighted collection. Negative elements will be ignored. 1Map function over weighted collection elements.         None None@Synchronize two datasets, taking disamb tags from the first one 2 and the rest of information form the second one. L In case of differences in token-level segmentation, reference segmentation G (token-level) is assumed. Otherwise, it would be difficult to choose  correct disamb tags. JIf both arguments contain only one segment, insert disamb interpretations L from the first segment into the second segment. Otherwise, the first list ) of segments will be returned unchanged. Align two lists of segments. DFind the shortest, length-matching prefixes in the two input lists. NoneIAn analyser performs word-level segmentation and morphological analysis. Reanalyse sentence. 0From the reference sentence the function takes:  Word-level segmentation  Chosen interpretations (tags) 1From the reanalysed sentence the function takes:  Potential interpretations Reanalyse paragraph.     None$!EConfiguration of the schema. All configuration elements specify the L range over which a particular observation type should be taken on account.  For example, the  [-1, 0, 2]- range means that observations of particular 2 type will be extracted with respect to previous (k - 1 ), current (k)  and after the next (k + 2-) positions when identifying the observation  set for position k in the input sentence. #The 8 schema block. $The 9 schema block. %The :' schema block. The first list of ints ! represents lengths of prefixes. &The ;' schema block. The first list of ints ! represents lengths of suffixes. 'The < schema block. (The = schema block. )The > schema block. *The ? schema block. + Maybe entry. ,Body of configuration entry. .&Range argument for the schema block. /1When true, the entry is used only for oov words. 0+Additional arguments for the schema block. 1>A block is a chunk of the Ox computation performed within the F context of the sentence and the list of absolute sentence positions. 1Record structure of the basic observation types. 2?A schema is a block of the Ox computation performed within the = context of the sentence and the absolute sentence position. 3CThe Ox monad specialized to word token type and text observations. 4@An observation consist of an index (of list type) and an actual  observation value. 5A dummy schema block. 6HSequence the list of schemas (or blocks) and discard individual values. Construct the  structure given the sentence. 7+Transform a block to a schema depending on * * A list of relative sentence positions, 3 * A boolean value; if true, the block computation 4 will be performed only on positions where an OOV  word resides. 8+Orthographic form at the current position. 9+Orthographic form at the current position. :.List of lowercased prefixes of given lengths. ;.List of lowercased suffixes of given lengths. <Shape of the word. =Shape of the word. >Packed shape of the word. ?Packed shape of the word. @!Entry with additional arguemnts. A*Plain entry with no additional arugments. B.Null configuration of the observation schema. C-Build the schema based on the configuration. D:Use the schema to extract observations from the sentence. -!"#$%&'()*+,-./0123456789:;<=>?@ABCD$!"#$%&'()*+,-./0123456789:;<=>?@ABCD$43256D,-./0+A@!"#$%&'()*BC1789:;<=>?! "#$%&'()*+,-./0123456789:;<=>?@ABCDNoneETraining configuration. IA guessing model. 0Schematize the input sentence with according to schema rules. M Determine k5 most probable labels for each word in the sentence. N+Insert guessing results into the sentence. OCombine M with N. PTrain guesser. &Schematized data from the plain file. EFGHIJKLMNOPTraining configuration Training data Maybe evaluation data EFGHIJKLMNOP IJKLMNOEFGHP EFGHIJKLMNOPNone QTraining configuration. VA disambiguation model. 0Schematize the input sentence with according to schema rules. 2Unsplit the complex tag (assuming, that it is one & of the interpretations of the word). [*Perform context-sensitive disambiguation. \1Insert disambiguation results into the sentence. ]Combine [ with \. ^Train disamb model. &Schematized data from the plain file. QRSTUVWXYZ[\]^Training configuration Training data Maybe evaluation data Resultant model QRSTUVWXYZ[\]^VWXYZ[\]QRSTU^ QRSTUVWXYZ[\]^None_Concraft data. eASave model in a file. Data is compressed using the gzip format. fLoad model from a file. g@Tag sentence using the model. In your code you should probably C use your analysis function, translate results into a container of  ences, evaluate tagSent on each sentence and embed the = tagging results into morphosyntactic structure of your own. h*Train guessing and disambiguation models. BStore dataset on a disk and run a handler on a list which is read ? lazily from the disk. A temporary file will be automatically $ deleted after the handler is done.  Similar to  but on a  dataset. BStore dataset on a disk and run a handler on a list which is read ? lazily from the disk. A temporary file will be automatically $ deleted after the handler is done. _`abcdefghTagset Analysis function %Numer of guessed tags for each word &Guessing model training configuration ,Disambiguation model training configuration Training data Maybe evaluation data  Template for  Input dataset Handler _`abcdefgh _`abcdefgh _`abcdefgh     !"#$%&'(()*+,-./01223456789:;<=>?@ABCDEFGHIJJKLMMNOPQRSJJTKLUUVNOWQXSYYZ[\W]^_S`abcdefgh9ijklS_mnodpqrstuvwxyzh{9|}~ IIconcraft-0.5.0NLP.Concraft.DisambNLP.Concraft.MorphosyntaxNLP.Concraft.AnalysisNLP.Concraft.SchemaNLP.Concraft.Guess NLP.ConcraftNLP.Concraft.Disamb.PositionalNLP.Concraft.Disamb.TieredNLP.Concraft.Format.TempNLP.Concraft.Morphosyntax.AlignAtomposattsTierwithPoswithAttsCRFWMapunWMapSentOsegsorigSentWordorthoovSegwordtagsmapSeg interpsSetinterpsmapSentmapSentOmkWMapmapWMapAnalyse reAnaSentreAnaPar SchemaConforthClowOrthC lowPrefixesC lowSuffixesCknownCshapeCpackedC begPackedCEntryBodyrangeoovOnlyargsBlockSchemaOxObvoid sequenceS_ fromBlockorthBlowOrthB lowPrefixesB lowSuffixesBknownBshapeBpackedB begPackedB entryWithentrynullConffromConf schematize TrainConf schemaConfTsgdArgsTGuesser schemaConfcrfguessinclude guessSenttraintiersTDisambtiersdisamb disambSentConcrafttagsetguessNumguesser saveModel loadModeltagselectsplit $fBinaryAtom $fBinaryTierFeatMapTransMap CodecDataFeatLbfeatGencodecdummy codecSpec numOfLayers codecDatamodel transMapsotherMapOFeatobTFeat1TFeat2TFeat3x1x2x3lnunLbunObputIgetIobLenslbLens!?ghc-prim GHC.TypesIO $fBinaryCRF$fFeatMapFeatMapFeat$fBinaryFeatMap $fBinaryFeat $fWordSeg $fFromJSONSeg $fToJSONSeg encodePar decodeParwriteParreadParsync moveDisambalignmatchlazyMapMBaseObmkBaseOblowOrthmkArg0mkArg1$fBinarySchemaConf $fBinaryBodyschemed$fBinaryGuesserunSplit$fBinaryDisambwithTemp withTemp'base Data.MaybeMaybe modelVersiontemporary-1.1.2.4System.IO.Temp withTempFile$fBinaryConcraft