`wangchanberta-base-wiki-spm`

Pretrained RoBERTa BASE model on Thai Wikipedia corpus.

Model description

The architecture of the pretrained model is based on RoBERTa [Liu et al., 2019].

Intended uses & limitations

You can use the pretrained model for masked language modeling (i.e. predicting a mask token in the input text). In addition, we also provide finetuned models for multiclass/multilabel text classification and token classification task.

Multiclass text classification

wisesight_sentiment

4-class text classification task (positive, neutral, negative, and question) based on social media posts and tweets.
wongnai_reivews

Users' review rating classification task (scale is ranging from 1 to 5)
generated_reviews_enth : (review_star as label)

Generated users' review rating classification task (scale is ranging from 1 to 5).

Multilabel text classification

prachathai67k

Thai topic classification with 12 labels based on news article corpus from prachathai.com. The detail is described in this page.

Token classification

thainer

Named-entity recognition tagging with 13 named-entities as descibed in this page.
lst20 : NER NER and POS tagging

Named-entity recognition tagging with 10 named-entities and Part-of-Speech tagging with 16 tags as descibed in this page.

How to use

Training data

wangchanberta-base-wiki-spm model was pretrained on Thai Wikipedia. Specifically we use the Wikipedia dump articles on 20 August 2020 (dumps.wikimedia.org/thwiki/20200820/). We opt out lists, and tables.

Preprocessing

Texts are preprocessed with the following rules:

Replace non-breaking space, zero-width non-breaking space, and soft hyphen with spaces.
Remove an empty parenthesis that occur right after the title of the first paragraph.
Replace spaces wtth <_>.

Regarding the vocabulary, we use subword token trained with SentencePice library on the training set of Thai Wikipedia corpus. The total number of subword tokens is 24,000.

We sample sentences contigously to have the length of at most 512 tokens. For some sentences that overlap the boundary of 512 tokens, we split such sentence with an additional token as document separator. This is the same approach as proposed by [Liu et al., 2019] (called "FULL-SENTENCES").

The details of the masking procedure for each sentence are the following:

Train/Val/Test splits

We split sequencially 944,782 sentences for training set, 24,863 sentences for validation set and 24,862 sentences for test set.