######## This is the default tagger.properties file
######## This file is used for training and testing only,
######## The configuration for tagging is directly tuned in the descriptor "HmmTagger.xml"

###########  ONLY FOR TRAINING #####################################################


####### FILE OR DIRECTORY CONTAINING TRAINING CORPUS:
####### can be in specified either as an absolute or as a relative path
####### e.g. FILE = ../../tueba_tigerFormat.txt or FILE = C:/Data/tueba.txt
FILE = 


##########################  BOTH FOR TRAINING AND EVALUATION  ################################

######## THESE ARE THE DEFAULT MODEL FILES FOR GERMAN AND ENGLISH
######## You can either uncomment one of them, if you want to replace given models with your own one,

#MODEL_FILE = resources/german/TuebaModel.dat
MODEL_FILE = resources/english/BrownModel.dat

######## or specify a completely different name
# MODEL_FILE = 

######## If mapping of tags is desired, uncomment the following
DO_MAPPING = true


#######	EXAMPLES OF MAPPING CLASSES 

## Basic mapping for the Brown corpus (nltk distribution) tagset: to get 93 tags out of 473
MAPPING = org.apache.uima.examples.tagger.trainAndTest.TagMappingBrown

## Basic mapping for STTS tagset: from 54 tags onto the basic ca. 15 classes plus punctuation
#MAPPING = org.apache.uima.examples.tagger.trainAndTest.GrobMappingTueba

## If you implement your own mapping, you should specify here in the same manner as above a java-path to the class  
#MAPPING = 

######## If corpus is in a different format and cannot be read with the provided READERS,
######## you should specify here a java-path to the class (s. examples below) 

#CORPUS_READER = org.apache.uima.examples.tagger.trainAndTest.TT_FormatReader
CORPUS_READER = org.apache.uima.examples.tagger.trainAndTest.BrownReader

#CORPUS_READER = 

#################      ONLY FOR EVALUATION   ###############################

######### GOLD STANDARD CORPUS FILE: 
######### can be specified as an absolute or as a relative path 
##e.g. GOLD_STANDARD = ../../tueba_tigerFormat.txt or GOLD_STANDARD = C:/Data/tueba.txt

GOLD_STANDARD = 

######### Here we specify whether one intends to test a bi- or a trigram model (default is a trigram model)  
N=3