beto-sarcasmo-bs32-ep6
This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4817
- F1 Macro: 0.9125
- F1 Weighted: 0.9167
- Accuracy: 0.9168
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: 16
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 0.1
- num_epochs: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Weighted | Accuracy |
|---|---|---|---|---|---|---|
| 0.6315 | 1.0 | 478 | 0.2967 | 0.8736 | 0.8782 | 0.8772 |
| 0.3828 | 2.0 | 956 | 0.2392 | 0.9030 | 0.9077 | 0.9079 |
| 0.2742 | 3.0 | 1434 | 0.2892 | 0.9094 | 0.9137 | 0.9139 |
| 0.1957 | 4.0 | 1912 | 0.3759 | 0.9120 | 0.9159 | 0.9157 |
| 0.1127 | 5.0 | 2390 | 0.4284 | 0.9140 | 0.9184 | 0.9188 |
| 0.0526 | 6.0 | 2868 | 0.4817 | 0.9125 | 0.9167 | 0.9168 |
Framework versions
- Transformers 5.12.1
- Pytorch 2.11.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2