Thanks for the reply. Will try these.
On Sep 26, 2016 1:45 PM, "Hieu Hoang" wrote:
> yes, 1500 sentences is too small. Other problems could be
>
> 1. You haven't tokenized the data
>
> 2. Your data files have different encoding. Moses expects files to be
> encoded in
yes, 1500 sentences is too small. Other problems could be
1. You haven't tokenized the data
2. Your data files have different encoding. Moses expects files to be
encoded in utf8
___
Moses-support mailing list
Moses-support@mit.edu
Dear all,
I am using Moses to translate Amharic text to English based on the Moses
guideline I found on http://www.statmt.org/moses/?n=Development.GetStarted.
I have trained the system with a parallel corpus of 1500 sentences of each
language.
I am wondering if this is because of the amount of