IndoJavE: BERT-base
About
This is a pre-trained masked language model for code-mixed Indonesian-Javanese-English tweets data. This model is trained based on BERT model utilizing Hugging Face's Transformers library.
Pre-training Data
The Twitter data is collected from January 2022 until January 2023. The tweets are collected using 8698 random keyword phrases. To make sure the retrieved data are code-mixed, we use keyword phrases that contain code-mixed Indonesian, Javanese, or English words. The following are few examples of the keyword phrases:
- travelling terus
- proud koncoku
- great kalian semua
- chattingane ilang
- baru aja launching
We acquire 40,788,384 raw tweets. We apply first stage pre-processing tasks such as:
- remove duplicate tweets,
- remove tweets with token length less than 5,
- remove multiple space,
- convert emoticon,
- convert all tweets to lower case.
After the first stage pre-processing, we obtain 17,385,773 tweets. In the second stage pre-processing, we do the following pre-processing tasks:
- split the tweets into sentences,
- remove sentences with token length less than 4,
- convert ‘@username’ to ‘@USER’,
- convert URL to HTTPURL.
Finally, we have 28,121,693 sentences for the training process. This pretraining data will not be opened to public due to Twitter policy.
Model
| Model name | Architecture | Size of training data | Size of validation data |
|---|---|---|---|
indojave-codemixed-bert-base | BERT | 2.24 GB of text | 249 MB of text |
Evaluation Results
We train the data with 3 epochs and total steps of 296K for 12 days. The following are the results obtained from the training:
| train loss | eval loss | eval perplexity |
|---|---|---|
| 3.5057 | 3.0559 | 21.2398 |
How to use
Load model and tokenizer
from transformers import AutoTokenizer, AutoModel
tokenizer = AutoTokenizer.from_pretrained("fathan/indojave-codemixed-bert-base")
model = AutoModel.from_pretrained("fathan/indojave-codemixed-bert-base")
Masked language model
from transformers import pipeline
pretrained_model = "fathan/indojave-codemixed-bert-base"
fill_mask = pipeline(
"fill-mask",
model=pretrained_model,
tokenizer=pretrained_model
)
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- 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
- num_epochs: 3.0
Framework versions
- Transformers 4.26.0
- Pytorch 1.12.0+cu102
- Datasets 2.9.0
- Tokenizers 0.12.1