úÎC“?¼6      !"#$%&'()*+,-./012345NoneDA NeDict is a map from forms to NE types. Each NE may be annotated  with multiple types. .A Named Entity entry from the LMF dictionary. Orthographic form of the NE Type of the NE A type of named entity. A orthographic form. Is the form a multiword one? 3Construct the dictionary from the list of entries. >Remove dictionary entries which do not satisfy the predicate. !Save the dictionary in the file. #Load the dictionary from the file. Merge dictionary resources. 6Differentiate labels from separate dictionaries using  dictionary-unique prefixes. 6Map function over the set. 7Add integer prefix.  67   67None-Parse the dictionary to the list of entries. #Read the dictionary from the file. 89:;<89:;<None,Parse the NELexicon into a list of entries. #Read the dictionary from the file. ==None>3Make dictionary consisting only from one word NEs. <Parse the PNEG dictionary and save it in a binary form into  the output file. >Parse the NELexicon, merge it with the PoliMorf and serialize  into a binary, DAWG form. >Path to PNEG in the LMF format  Output file Path to NELexicon Path to PoliMorf  Output file  >NoneAA label is created by encoding the named entity forest using the  IOB method. @An observation consist of an index (of list type) and an actual  observation value. A named entity. A word. NoneEConfiguration 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 & schema block The ' schema block The ( schema block The ) schema block The * schema block The + schema block >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. !?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. #A dummy schema block. $<Sequence the list of schemas and discard individual values. @Construct the ? structure given the sentence. %;Transform the block to the schema dependent on the list of  relative sentence positions. &9Orthographic observations determined with respect to the  list of relative positions. '8Lemma substitute determined with respect to the list of  relative positions. (>Shape and packed shape determined with respect to the list of  relative positions. )GShape pairs determined with respect to the list of relative positions. *8Several suffixes determined with respect to the list of  relative positions. +?Plain dictionary search determined with respect to the list of  relative positions. ,3Default configuration for Nerf observation schema. --Build the schema based on the configuration. .:Use the schema to extract observations from the sentence.  ?ABC!"#$@%&'()*+,Path to  in a binary form DE-.F !"#$%&'()*+,-."!#$.,- %&'()*+ ?ABC!"#$@%&'()*+,DE-.FNone/KA Nerf consists of the observation schema configuration and the CRF model. 3AShow results of observation extraction on the input ENAMEX file. 43Train Nerf on the input data using the SGD method. 57Perform named entity recognition (NER) using the Nerf. /012GHI34 Args for SGD !Observation schema configuration Train data (ENAMEX) Maybe eval data (ENAMEX) $Nerf with resulting codec and model 5J /012345/012453/012GHI345JK       !"#$%&'()*+,-./01234556789:;<=>?@ABCDEDFGHIJKLMNO nerf-0.1.0NLP.Nerf.Dict.BaseNLP.Nerf.Dict.PNEGNLP.Nerf.Dict.NELexicon NLP.Nerf.DictNLP.Nerf.TypesNLP.Nerf.SchemaNLP.NerfNeDictEntryneOrthneTypeNeTypeForm isMultiWordmkDictsiftDictsaveDictloadDictmergediff parsePNEGreadPNEGparseNELexicon readNELexicon preparePNEGprepareNELexiconLbObNEWord SchemaCfgorthClemmaCshapeC shapePairCsuffixCdictCBlockSchemaOxvoid sequenceS_ fromBlockorthSlemmaSshapeS shapePairSsuffixSsearchS defaultCfgfromCfg schematizeNerf schemaCfgcrftryOxtrainnermapS addPrefixlmfP lexEntryPwordPsensePfeatP parseLinemkDictW1BaseObmkBaseOborthlowOrthmkBasicSmkDictS$fBinarySchemaCfgflattenreadFlatdrawSent $fBinaryNerf