bert_adaptation_martin_fierro
This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 3.6952
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: 2e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 5.3685 | 1.0 | 58 | 4.8544 |
| 4.3651 | 2.0 | 116 | 4.4985 |
| 3.8868 | 3.0 | 174 | 3.7412 |
| 3.7072 | 4.0 | 232 | 3.7141 |
| 3.7598 | 5.0 | 290 | 3.8915 |
| 3.5408 | 6.0 | 348 | 3.9818 |
| 3.3939 | 7.0 | 406 | 4.0336 |
| 3.4642 | 8.0 | 464 | 3.8640 |
| 3.2745 | 9.0 | 522 | 3.9118 |
| 3.177 | 10.0 | 580 | 3.8540 |
| 3.2224 | 11.0 | 638 | 3.8703 |
| 3.1109 | 12.0 | 696 | 3.7848 |
| 2.9985 | 13.0 | 754 | 3.5726 |
| 3.0286 | 14.0 | 812 | 3.8872 |
| 3.0456 | 15.0 | 870 | 3.8465 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu118
- Datasets 2.15.0
- Tokenizers 0.15.0