<!-- This model card has been generated automatically according to the information Keras had access to. You should probably proofread and complete it, then remove this comment. -->
silviacamplani/distilbert-finetuned-ner-ai
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.8962
- Validation Loss: 0.9088
- Train Precision: 0.3895
- Train Recall: 0.3901
- Train F1: 0.3898
- Train Accuracy: 0.7558
- Epoch: 9
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: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 1e-05, 'decay_steps': 350, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
- training_precision: mixed_float16
Training results
| Train Loss | Validation Loss | Train Precision | Train Recall | Train F1 | Train Accuracy | Epoch |
|---|---|---|---|---|---|---|
| 2.5761 | 1.7934 | 0.0 | 0.0 | 0.0 | 0.6480 | 0 |
| 1.7098 | 1.5860 | 0.0 | 0.0 | 0.0 | 0.6480 | 1 |
| 1.4692 | 1.3213 | 0.0 | 0.0 | 0.0 | 0.6480 | 2 |
| 1.2755 | 1.1859 | 0.1154 | 0.0460 | 0.0658 | 0.6789 | 3 |
| 1.1561 | 1.0921 | 0.2878 | 0.2010 | 0.2367 | 0.7192 | 4 |
| 1.0652 | 1.0170 | 0.3250 | 0.2862 | 0.3043 | 0.7354 | 5 |
| 0.9936 | 0.9649 | 0.3489 | 0.3305 | 0.3395 | 0.7462 | 6 |
| 0.9442 | 0.9340 | 0.3845 | 0.3799 | 0.3822 | 0.7549 | 7 |
| 0.9097 | 0.9168 | 0.3866 | 0.3748 | 0.3806 | 0.7556 | 8 |
| 0.8962 | 0.9088 | 0.3895 | 0.3901 | 0.3898 | 0.7558 | 9 |
Framework versions
- Transformers 4.20.1
- TensorFlow 2.6.4
- Datasets 2.1.0
- Tokenizers 0.12.1
