Downloads · 30 days
88
10% of all-time downloads
w11wo/javanese-roberta-small-imdb
javanese-roberta-small-imdb is a fill-mask model from w11wo. Use it when you need the model to fill a missing word. It is set up for transformers. The card lists the license as mit.
Javanese RoBERTa Small IMDB is a masked language model based on the RoBERTa model. It was trained on Javanese IMDB movie reviews.
Downloads · 30 days
88
10% of all-time downloads
All-time downloads
863
Public
Parameters
125M
2.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.h5655 MB · 30%
How the weights are stored.
F32125M · 100%
From the Hugging Face model README
Javanese RoBERTa Small IMDB is a masked language model based on the RoBERTa model. It was trained on Javanese IMDB movie reviews.
The model was originally the pretrained Javanese RoBERTa Small model and is later fine-tuned on the Javanese IMDB movie review dataset. It achieved a perplexity of 20.83 on the validation dataset. Many of the techniques used are based on a Hugging Face tutorial notebook written by Sylvain Gugger.
Hugging Face's Trainer class from the Transformers library was used to train the model. PyTorch was used as the backend framework during training, but the model remains compatible with TensorFlow nonetheless.
| Model | #params | Arch. | Training/Validation data (text) |
|---|---|---|---|
javanese-roberta-small-imdb | 124M | RoBERTa Small | Javanese IMDB (47.5 MB of text) |
The model was trained for 5 epochs and the following is the final result once the training ended.
| train loss | valid loss | perplexity | total time |
|---|---|---|---|
| 3.140 | 3.036 | 20.83 | 2:59:28 |
from transformers import pipeline
pretrained_name = "w11wo/javanese-roberta-small-imdb"
fill_mask = pipeline(
"fill-mask",
model=pretrained_name,
tokenizer=pretrained_name
)
fill_mask("Aku mangan sate ing <mask> bareng konco-konco")
from transformers import RobertaModel, RobertaTokenizerFast
pretrained_name = "w11wo/javanese-roberta-small-imdb"
model = RobertaModel.from_pretrained(pretrained_name)
tokenizer = RobertaTokenizerFast.from_pretrained(pretrained_name)
prompt = "Indonesia minangka negara gedhe."
encoded_input = tokenizer(prompt, return_tensors='pt')
output = model(**encoded_input)
Do consider the biases which came from the IMDB review that may be carried over into the results of this model.
Javanese RoBERTa Small was trained and evaluated by Wilson Wongso. All computation and development are done on Google Colaboratory using their free GPU access.
If you use any of our models in your research, please cite:
@inproceedings{wongso2021causal,
title={Causal and Masked Language Modeling of Javanese Language using Transformer-based Architectures},
author={Wongso, Wilson and Setiawan, David Samuel and Suhartono, Derwin},
booktitle={2021 International Conference on Advanced Computer Science and Information Systems (ICACSIS)},
pages={1--7},
year={2021},
organization={IEEE}
}