This week I trimmed more data sets for an older version of training data since it also had testing data which the newer sets do not. This will allow us to see which one of the data sets give the best results. This week I also began installing programs to help with the next step where I will take the data that was trimmed on sentence length and trim it further based on things like what parts of speech are next to each other. From there I should be able to better group languages to prepare for the surprise languages.
0 Comments
|
|