Simple, standalone python classes for training statistical language models using several popular smoothing methods.
Also includes a script for evaluating LM sentence probabilities and retrieving individual NGram probabilities.
$ ./SimpleModKN.py -t train.corpus -o 3
Sentence:
$ ./evaluate-lm.py -m test.arpa -s "a b c d"
NGram:
$ ./evaluate-lm.py -m test.arpa -s "a b c" -g