---
license: cc
---
# Bert2gpt (Encoder-Decoder) on Liputan6 100k dataset
Dataset source: https://huggingface.co/datasets/fajrikoto/id_liputan6 <br>
Model used for Fine Tuning (Encoder):<br>
https://huggingface.co/cahya/bert-base-indonesian-1.5G <br>
Model used for Fine Tuning (Decoder):<br>
https://huggingface.co/cahya/gpt2-small-indonesian-522M <br><br>
Trained on 1x3090 @ 8 epoch (EarlyStopping Callbacks)

Train logs, metrics, and params: https://wandb.ai/willy030125/huggingface/runs/9nt3z9dh <br>
https://www.comet.com/willy030125/huggingface/404a9a8abbd84ed5931ff944746df83c <br>
Eval results and Perplexity: <a href="https://huggingface.co/Willy030125/Bert2gpt_Liputan6_100k_8epoch/blob/main/eval_results.json">eval_results.json</a><br>

Usage:
```python
from transformers import BertTokenizer, GPT2Tokenizer, EncoderDecoderModel
encoder_tokenizer = BertTokenizer.from_pretrained("cahya/bert-base-indonesian-1.5G")
decoder_tokenizer = GPT2Tokenizer.from_pretrained("cahya/gpt2-small-indonesian-522M")
model = EncoderDecoderModel.from_pretrained("Willy030125/Bert2gpt_Liputan6_100k_8epoch")
```