#############################################################################
## Copyright (c) 1996, Carnegie Mellon University, Cambridge University,
## Ronald Rosenfeld and Philip Clarkson
#############################################################################
=============================================================================
===============  This file was produced by the CMU-Cambridge  ===============
===============     Statistical Language Modeling Toolkit     ===============
=============================================================================
This is a 1-gram language model, based on a vocabulary of 28 words,
  which begins "<s>", "a", "b"...
This is an OPEN-vocabulary model (type 1)
  (OOVs were mapped to UNK, which is treated as any other vocabulary word)
Good-Turing discounting was applied.
1-gram frequency of frequency : 4 
1-gram discounting ratios : 0.67 
This file is in the ARPA-standard format introduced by Doug Paul.

p(wd3|wd1,wd2)= if(trigram exists)           p_3(wd1,wd2,wd3)
                else if(bigram w1,w2 exists) bo_wt_2(w1,w2)*p(wd3|wd2)
                else                         p(wd3|w2)

p(wd2|wd1)= if(bigram exists) p_2(wd1,wd2)
            else              bo_wt_1(wd1)*p_1(wd2)

All probs and back-off weights (bo_wt) are given in log10 form.

Data formats:

Beginning of data mark: \data\
ngram 1=nr            # number of 1-grams

\1-grams:
p_1     wd_1 

end of data mark: \end\

\data\
ngram 1=29

\1-grams:
-3.5087 <UNK> 0.0000
-98.9999 <s> 0.0000
-1.1422 a 0.0000
-1.4032 b 0.0000
-1.6336 c 0.0000
-1.4937 d 0.0000
-1.0272 e 0.0000
-1.9176 f 0.0000
-1.4518 g 0.0000
-1.2873 h 0.0000
-1.1834 i 0.0000
-2.1864 j 0.0000
-1.8011 k 0.0000
-1.2951 l 0.0000
-1.5922 m 0.0000
-1.2186 n 0.0000
-1.2288 o 0.0000
-1.4461 p 0.0000
-3.5087 q 0.0000
-1.0538 r 0.0000
-1.2288 s 0.0000
-1.0996 t 0.0000
-1.4349 u 0.0000
-1.8141 v 0.0000
-1.6891 w 0.0000
-3.5087 x 0.0000
-1.7885 y 0.0000
-2.4295 z 0.0000
-3.5087 </s> 0.0000

\end\