<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
fine-tuned-IndoNLI-Basic-with-indobert-base-uncased
This model is a fine-tuned version of indolem/indobert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7758
- Accuracy: 0.7761
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 1e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.06
- num_epochs: 16
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0858 | 1.0 | 161 | 0.9459 | 0.5230 |
| 0.8345 | 2.0 | 322 | 0.7373 | 0.6978 |
| 0.714 | 3.0 | 483 | 0.6561 | 0.7401 |
| 0.6385 | 4.0 | 644 | 0.6023 | 0.7679 |
| 0.541 | 5.0 | 805 | 0.5944 | 0.7756 |
| 0.4812 | 6.0 | 966 | 0.6186 | 0.7729 |
| 0.4285 | 7.0 | 1127 | 0.6241 | 0.7806 |
| 0.3847 | 8.0 | 1288 | 0.6587 | 0.7820 |
| 0.3295 | 9.0 | 1449 | 0.6775 | 0.7770 |
| 0.2998 | 10.0 | 1610 | 0.7086 | 0.7806 |
| 0.2585 | 11.0 | 1771 | 0.7179 | 0.7779 |
| 0.2409 | 12.0 | 1932 | 0.7355 | 0.7797 |
| 0.218 | 13.0 | 2093 | 0.7490 | 0.7729 |
| 0.204 | 14.0 | 2254 | 0.7683 | 0.7751 |
| 0.1981 | 15.0 | 2415 | 0.7748 | 0.7770 |
| 0.1956 | 16.0 | 2576 | 0.7758 | 0.7761 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cu117
- Datasets 2.2.0
- Tokenizers 0.13.2
