Closed amitli1 closed 1 year ago
Your corpus is too small to have the statistical regularities that modified Kneser-Ney smoothing expects when estimating discounts. If you want to kludigly make up some discounts, --discount_fallback
is there for you.
I have simple English file:
I'm running the following command:
lmplz -o 3 <myTest.txt >myTest.arpa
And getting error:
If I run it with
--discount_fallback
parameter - it works.