metadata
license: mit
tags:
- generated_from_keras_callback
model-index:
- name: tf-tpu/roberta-base-epochs-100
results: []
tf-tpu/roberta-base-epochs-100
This model is a fine-tuned version of roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 4.5963
- Train Accuracy: 0.0449
- Validation Loss: 4.5195
- Validation Accuracy: 0.0444
- Epoch: 14
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:
- optimizer: {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 0.0001, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0001, 'decay_steps': 55765, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 2935, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.001}
- training_precision: mixed_bfloat16
Training results
| Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
|---|---|---|---|---|
| 7.2121 | 0.0274 | 5.7188 | 0.0346 | 0 |
| 5.4335 | 0.0414 | 5.2266 | 0.0439 | 1 |
| 5.1579 | 0.0445 | 5.0625 | 0.0441 | 2 |
| 5.0231 | 0.0447 | 4.9453 | 0.0446 | 3 |
| 4.9323 | 0.0448 | 4.8633 | 0.0443 | 4 |
| 4.8672 | 0.0449 | 4.8789 | 0.0440 | 5 |
| 4.8200 | 0.0449 | 4.8164 | 0.0441 | 6 |
| 4.7841 | 0.0449 | 4.7734 | 0.0450 | 7 |
| 4.7546 | 0.0449 | 4.7539 | 0.0441 | 8 |
| 4.7288 | 0.0449 | 4.7305 | 0.0447 | 9 |
| 4.7084 | 0.0449 | 4.7422 | 0.0443 | 10 |
| 4.6884 | 0.0450 | 4.7148 | 0.0437 | 11 |
| 4.6764 | 0.0449 | 4.7070 | 0.0441 | 12 |
| 4.6637 | 0.0449 | 4.7227 | 0.0435 | 13 |
| 4.5963 | 0.0449 | 4.5195 | 0.0444 | 14 |
Framework versions
- Transformers 4.27.0.dev0
- TensorFlow 2.9.1
- Tokenizers 0.13.2