Add LST20 Part-Of-Speech tagger model

[wannaphong]

LST20 Corpus from National Electronics and Computer Technology Center, Thailand. It can download dataset from https://aiforthai.in.th/corpus.php.

Support Model

• Unigram Part-Of-Speech tagger
• Perceptron Part-Of-Speech tagger

and tag map for LST20 to Universal Dependencies.

Model train by Mr.Wannaphong Phatthiyaphaibun

Model License : CC-0

Code

• Model File : https://github.com/PyThaiNLP/pythainlp-corpus/releases/tag/lst20-v0.2
• Train script : https://github.com/PyThaiNLP/pythainlp_notebook/blob/master/postag/train_lst20_pythainlp.ipynb
• Google Colab : https://colab.research.google.com/drive/1Fpp1iCdx1mGcnAOTnU2O7QwY5IsqHHtW?usp=sharing

TODO

• document

[pep8speaks]

Hello @wannaphong! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:

There are currently no PEP 8 issues detected in this Pull Request. Cheers! 🍻

Comment last updated at 2020-08-20 10:32:19 UTC

[coveralls]

Coverage increased (+0.2%) to 95.222% when pulling 557dd7c on add-LST20-postag into e6e01c5 on dev.

[wannaphong]

I doing train new model. I will combine eval with train.

[wannaphong]

I doing train new model. I will combine eval with train.

Done

[p16i]

It would be nice if we have some comparisons between these taggers and also the existing ones we have in PyThaiNLP.

[bact]

Looks good

[bact]

I refactor some of the code, move tagger-related stuffs away from the corpus data files (lst20.py and unigram.py) and put them to tagger files (perceptron.py and unigram.py)