📄 an4_spelling.unigram.lm
字号:
############################################################################### 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\
⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -