<!-- 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. -->
ratish/DBERT_ZS_Desc_MAKE_v1.4.1
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: 0.0478
- Validation Loss: 0.0497
- Train Accuracy: 1.0
- Epoch: 25
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': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 3060, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
- training_precision: float32
Training results
Train Loss | Validation Loss | Train Accuracy | Epoch |
---|---|---|---|
1.4853 | 1.3110 | 0.8889 | 0 |
1.2406 | 0.9999 | 0.8889 | 1 |
0.9564 | 0.7717 | 0.8889 | 2 |
0.7727 | 0.6130 | 0.8889 | 3 |
0.5895 | 0.4969 | 0.8889 | 4 |
0.4935 | 0.4101 | 0.8889 | 5 |
0.3936 | 0.3439 | 0.8889 | 6 |
0.3186 | 0.2861 | 0.8889 | 7 |
0.2737 | 0.2477 | 0.8889 | 8 |
0.2171 | 0.2156 | 1.0 | 9 |
0.1996 | 0.1904 | 1.0 | 10 |
0.1732 | 0.1709 | 1.0 | 11 |
0.1494 | 0.1504 | 1.0 | 12 |
0.1343 | 0.1323 | 1.0 | 13 |
0.1065 | 0.1196 | 1.0 | 14 |
0.1000 | 0.1083 | 1.0 | 15 |
0.0961 | 0.0992 | 1.0 | 16 |
0.0922 | 0.0904 | 1.0 | 17 |
0.0796 | 0.0833 | 1.0 | 18 |
0.0666 | 0.0764 | 1.0 | 19 |
0.0640 | 0.0713 | 1.0 | 20 |
0.0649 | 0.0675 | 1.0 | 21 |
0.0612 | 0.0626 | 1.0 | 22 |
0.0539 | 0.0579 | 1.0 | 23 |
0.0512 | 0.0532 | 1.0 | 24 |
0.0478 | 0.0497 | 1.0 | 25 |
Framework versions
- Transformers 4.28.1
- TensorFlow 2.12.0
- Datasets 2.11.0
- Tokenizers 0.13.3