Update sentenceseg_crfcut.model

[wannaphong]

What does this changes

Retrained sentenceseg_crfcut.model for PyThaiNLP 2.4

What was wrong

After #557 & #562, We needs to retrained all model in PyThaiNLP before we release PyThaiNLP 2.4 because dict & newmm has update.

How this fixes it

Description of how the changes fix the issue.

Fixes #581

Your checklist for this pull request

🚨Please review the guidelines for contributing to this repository.

• Passed code styles and structures
• Passed code linting checks and unit test

[wannaphong]

Log for retrained: https://github.com/PyThaiNLP/pythainlp_notebook/blob/master/sentenceseg_3dataset.ipynb

[wannaphong]

It is same f1-score.

[wannaphong]

before 4.32 MB after 3.9 MB ...

@cstorm125 @bact What are you think about it?

[coveralls]

Coverage decreased (-0.05%) to 95.778% when pulling 94d24b3 on add-retrained-crfcut-model into 259daf1 on dev.

[bact]

before 4.32 MB after 3.9 MB ...

@cstorm125 @bact What are you think about it?

Maybe related to the smaller vocab size?
words_th.txt is smaller in new version of PyThaiNLP after removing misspelled words.

Guessing.

[wannaphong]

before 4.32 MB after 3.9 MB ...
@cstorm125 @bact What are you think about it?

Maybe related to the smaller vocab size?
words_th.txt is smaller in new version of PyThaiNLP after removing misspelled words.

Guessing.

Same.