|
--- |
|
license: cc |
|
--- |
|
# Bert2Bert (Encoder-Decoder) on Liputan6 100k dataset |
|
Dataset source: https://huggingface.co/datasets/fajrikoto/id_liputan6 <br> |
|
Model used for Fine Tuning (Encoder-Decoder):<br> |
|
https://huggingface.co/indolem/indobert-base-uncased <br><br> |
|
Trained on 1x3090 @ 8 epoch (EarlyStopping Callbacks) |
|
|
|
Train logs, metrics, and params: https://wandb.ai/willy030125/huggingface/runs/2qk3jtic <br> |
|
https://www.comet.com/willy030125/huggingface/560ed6ccde1240c8b4401918fd27253a <br> |
|
Eval results and Perplexity: <a href="https://huggingface.co/Willy030125/Bert2Bert_Liputan6_100k_10epoch_IndoBERT/blob/main/eval_results.json">eval_results.json</a><br> |
|
|
|
Usage: |
|
```python |
|
from transformers import AutoTokenizer, EncoderDecoderModel |
|
tokenizer = AutoTokenizer.from_pretrained("Willy030125/Bert2Bert_Liputan6_100k_10epoch_IndoBERT") |
|
model = EncoderDecoderModel.from_pretrained("Willy030125/Bert2Bert_Liputan6_100k_10epoch_IndoBERT") |
|
``` |
