problem with nltk.lm module

[Monali Mavani]

Hello all,

I am facing a problem with nltk.lm module.

After running the following code multiple times, I get different results:

lm.fit(train, vocab)

len(lm.vocab)

in output sometimes I get 0 and sometimes a number for the same text and code.

print(lm.counts)

Outout: <NgramCounter with 1 ngram orders and 0 ngrams>

Even with learned vocab, ngrams are not generated.

I am attaching the code with the output.

I do not understand this behavior - different results at different times.

Thanks in advance. I am new to this group. Any help will be appreciated.

Monali