metadata
license: mit
base_model: roberta-base
tags:
- generated_from_keras_callback
model-index:
- name: Ryukijano/masked-lm-tpu
results: []
Ryukijano/masked-lm-tpu
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: 8.1956
- Train Accuracy: 0.0231
- Validation Loss: 8.1208
- Validation Accuracy: 0.0222
- Epoch: 11
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': 111625, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 5875, '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: float32
Training results
| Train Loss | Train Accuracy | Validation Loss | Validation Accuracy | Epoch |
|---|---|---|---|---|
| 10.2437 | 0.0000 | 10.1909 | 0.0000 | 0 |
| 10.1151 | 0.0001 | 9.9763 | 0.0016 | 1 |
| 9.8665 | 0.0107 | 9.6535 | 0.0215 | 2 |
| 9.5331 | 0.0230 | 9.2992 | 0.0223 | 3 |
| 9.2000 | 0.0231 | 8.9944 | 0.0222 | 4 |
| 8.9195 | 0.0229 | 8.7450 | 0.0224 | 5 |
| 8.6997 | 0.0231 | 8.6124 | 0.0219 | 6 |
| 8.5689 | 0.0229 | 8.4904 | 0.0222 | 7 |
| 8.4525 | 0.0230 | 8.3865 | 0.0223 | 8 |
| 8.3594 | 0.0230 | 8.3069 | 0.0221 | 9 |
| 8.2662 | 0.0231 | 8.2092 | 0.0224 | 10 |
| 8.1956 | 0.0231 | 8.1208 | 0.0222 | 11 |
Framework versions
- Transformers 4.32.1
- TensorFlow 2.12.0
- Tokenizers 0.13.3