indobart-v2
This model is a fine-tuned version of indobenchmark/indobart-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4389
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
No log | 0.18 | 100 | 1.3635 |
No log | 0.36 | 200 | 0.9978 |
No log | 0.54 | 300 | 0.9189 |
No log | 0.72 | 400 | 0.8184 |
1.3304 | 0.9 | 500 | 0.7635 |
1.3304 | 1.08 | 600 | 0.7604 |
1.3304 | 1.27 | 700 | 0.7000 |
1.3304 | 1.45 | 800 | 0.6662 |
1.3304 | 1.63 | 900 | 0.6032 |
0.7574 | 1.81 | 1000 | 0.5859 |
0.7574 | 1.99 | 1100 | 0.5648 |
0.7574 | 2.17 | 1200 | 0.5342 |
0.7574 | 2.35 | 1300 | 0.5271 |
0.7574 | 2.53 | 1400 | 0.4957 |
0.527 | 2.71 | 1500 | 0.4817 |
0.527 | 2.89 | 1600 | 0.4804 |
0.527 | 3.07 | 1700 | 0.4790 |
0.527 | 3.25 | 1800 | 0.4694 |
0.527 | 3.44 | 1900 | 0.4727 |
0.4096 | 3.62 | 2000 | 0.4677 |
0.4096 | 3.8 | 2100 | 0.4580 |
0.4096 | 3.98 | 2200 | 0.4614 |
0.4096 | 4.16 | 2300 | 0.4650 |
0.4096 | 4.34 | 2400 | 0.4617 |
0.3097 | 4.52 | 2500 | 0.4496 |
0.3097 | 4.7 | 2600 | 0.4484 |
0.3097 | 4.88 | 2700 | 0.4579 |
0.3097 | 5.06 | 2800 | 0.4471 |
0.3097 | 5.24 | 2900 | 0.4431 |
0.2617 | 5.42 | 3000 | 0.4481 |
0.2617 | 5.61 | 3100 | 0.4393 |
0.2617 | 5.79 | 3200 | 0.4345 |
0.2617 | 5.97 | 3300 | 0.4347 |
0.2617 | 6.15 | 3400 | 0.4437 |
0.2101 | 6.33 | 3500 | 0.4491 |
0.2101 | 6.51 | 3600 | 0.4452 |
0.2101 | 6.69 | 3700 | 0.4447 |
0.2101 | 6.87 | 3800 | 0.4346 |
0.2101 | 7.05 | 3900 | 0.4458 |
0.1721 | 7.23 | 4000 | 0.4523 |
0.1721 | 7.41 | 4100 | 0.4454 |
0.1721 | 7.59 | 4200 | 0.4389 |
Framework versions
- Transformers 4.33.3
- Pytorch 2.0.1
- Datasets 2.14.5
- Tokenizers 0.13.3
- Downloads last month
- 1