pythainlp.wangchanberta

The pythainlp.wangchanberta module is built upon the WangchanBERTa base model, specifically the wangchanberta-base-att-spm-uncased model, as detailed in the paper by Lowphansirikul et al. [^Lowphansirikul_2021].

This base model is utilized for various natural language processing tasks in the Thai language, including named entity recognition, part-of-speech tagging, and subword tokenization.

If you intend to fine-tune the model or explore its capabilities further, please refer to the [thai2transformers repository](https://github.com/vistec-AI/thai2transformers).

Speed Benchmark

Function	Named Entity Recognition	Part of Speech
PyThaiNLP basic function	89.7 ms	312 ms
pythainlp.wangchanberta (CPU)	9.64 s	9.65 s
pythainlp.wangchanberta (GPU)	8.02 s	8 s

For a comprehensive performance benchmark, the following notebooks are available:

• PyThaiNLP basic function and pythainlp.wangchanberta CPU at Google Colab

• pythainlp.wangchanberta GPU

Modules

References

[^Lowphansirikul_2021] Lowphansirikul L, Polpanumas C, Jantrakulchai N, Nutanong S. WangchanBERTa: Pretraining transformer-based Thai Language Models. [ArXiv:2101.09635](http://arxiv.org/abs/2101.09635) [Internet]. 2021 Jan 23 [cited 2021 Feb 27].