---
license: mit
datasets:
- scb_mt_enth_2020
- oscar
- wikipedia
- best2009
language:
- th
library_name: transformers
---
# HoogBERTa

This repository includes the Thai pretrained language representation (HoogBERTa_base) and the fine-tuned model for multitask sequence labeling.  


# Documentation


## Prerequisite
Since we use subword-nmt BPE encoding, input needs to be pre-tokenize using [BEST](https://huggingface.co/datasets/best2009) standard before inputting into HoogBERTa 
```
pip install attacut
```

## Getting Start
To initialize the model from hub, use the following commands
```python
from transformers import AutoTokenizer, AutoModel
from attacut import tokenize

tokenizer = AutoTokenizer.from_pretrained("new5558/HoogBERTa")
model = AutoModel.from_pretrained("new5558/HoogBERTa")
```

To annotate POS, NE, and clause boundary, use the following commands
```

```

To extract token features, based on the RoBERTa architecture, use the following commands

```python
with torch.no_grad():
    model.eval()
    sentence = "วันที่ 12 มีนาคมนี้ ฉันจะไปเที่ยววัดพระแก้ว ที่กรุงเทพ"
    all_sent = []
    sentences = sentence.split(" ")
    for sent in sentences:
        all_sent.append(" ".join(tokenize(sent)).replace("_","[!und:]"))

    sentence = " _ ".join(all_sent)
    tokenized_text = tokenizer(sentence, return_tensors = 'pt')
    token_ids = tokenized_text['input_ids']
    features = model(**tokenized_text)
```

For batch processing,

```python
with torch.no_grad():
    model.eval()
    sentenceL = ["วันที่ 12 มีนาคมนี้","ฉันจะไปเที่ยววัดพระแก้ว ที่กรุงเทพ"]
    inputList = []
    for sentX in sentenceL:
        sentences = sentX.split(" ")
        all_sent = []
        for sent in sentences:
            all_sent.append(" ".join(tokenize(sent)).replace("_","[!und:]"))

        sentence = " _ ".join(all_sent)
        inputList.append(sentence)
    tokenized_text = tokenizer(inputList, padding = True, return_tensors = 'pt')
    token_ids = tokenized_text['input_ids']
    features = model(**tokenized_text)
```

To use HoogBERTa as an embedding layer, use

```python
with torch.no_grad():
  features = model(token_ids) # where token_ids is a tensor with type "long".
```


## Conversion Code
If you are interested in how to convert Fairseq and subword-nmt Roberta into Huggingface hub here is my code used to do the conversion and test for parity match: 
https://www.kaggle.com/norapatbuppodom/hoogberta-conversion


# Citation

Please cite as:

``` bibtex
@inproceedings{porkaew2021hoogberta,
  title = {HoogBERTa: Multi-task Sequence Labeling using Thai Pretrained Language Representation},
  author = {Peerachet Porkaew, Prachya Boonkwan and Thepchai Supnithi},
  booktitle = {The Joint International Symposium on Artificial Intelligence and Natural Language Processing (iSAI-NLP 2021)},
  year = {2021},
  address={Online}
}
```

Download full-text [PDF](https://drive.google.com/file/d/1hwdyIssR5U_knhPE2HJigrc0rlkqWeLF/view?usp=sharing)  
Check out the code on [Github](https://github.com/lstnlp/HoogBERTa)