metadata

license: apache-2.0
language:
  - ind
  - ace
  - ban
  - bjn
  - bug
  - gor
  - jav
  - min
  - msa
  - nia
  - sun
  - tet
language_bcp47:
  - jv-x-bms
datasets:
  - sabilmakbar/indo_wiki
  - acul3/KoPI-NLLB
  - uonlp/CulturaX
tags:
  - bert

NusaBERT Base

NusaBERT Base is a multilingual encoder-based language model based on the BERT architecture. We conducted continued pre-training on open-source corpora of sabilmakbar/indo_wiki, acul3/KoPI-NLLB, and uonlp/CulturaX. On a held-out subset of the corpus, our model achieved:

eval_accuracy: 0.6866
eval_loss: 1.4876
perplexity: 4.4266

This model was trained using the 🤗Transformers PyTorch framework. All training was done on an NVIDIA H100 GPU. LazarusNLP/NusaBERT-base is released under Apache 2.0 license.

Model Detail

Developed by: LazarusNLP
Finetuned from: IndoBERT base p1
Model type: Encoder-based BERT language model
Language(s): Indonesian, Acehnese, Balinese, Banjarese, Buginese, Gorontalo, Javanese, Banyumasan, Minangkabau, Malay, Nias, Sundanese, Tetum
License: Apache 2.0
Contact: LazarusNLP

Use in 🤗Transformers

from transformers import AutoTokenizer, AutoModelForMaskedLM

model_checkpoint = "LazarusNLP/NusaBERT-base"

tokenizer = AutoTokenizer.from_pretrained(model_checkpoint)
model = AutoModelForMaskedLM.from_pretrained(model_checkpoint)

Training Datasets

Around 16B tokens from the following corpora were used during pre-training.

Training Hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0003
train_batch_size: 256
eval_batch_size: 256
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 24000
training_steps: 500000

Framework versions

Transformers 4.37.2
Pytorch 2.2.0+cu118
Datasets 2.17.1
Tokenizers 0.15.1

Credits

NusaBERT Base is developed with love by:

Citation

@misc{wongso2024nusabert,
  title={NusaBERT: Teaching IndoBERT to be Multilingual and Multicultural}, 
  author={Wilson Wongso and David Samuel Setiawan and Steven Limcorn and Ananto Joyoadikusumo},
  year={2024},
  eprint={2403.01817},
  archivePrefix={arXiv},
  primaryClass={cs.CL}
}