<!-- 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. -->
train_from_raw_cv12_true_suppress__0015
This model is a fine-tuned version of openai/whisper-tiny on an unknown dataset. It achieves the following results on the evaluation set:
- Train Loss: 0.3311
- Train Accuracy: 0.0999
- Train Wermet: 3.3402
- Validation Loss: 0.3658
- Validation Accuracy: 0.0617
- Validation Wermet: 8.6094
- 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': 2e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0.01}
- training_precision: float32
Training results
| Train Loss | Train Accuracy | Train Wermet | Validation Loss | Validation Accuracy | Validation Wermet | Epoch |
|---|---|---|---|---|---|---|
| 2.3569 | 0.0439 | 2.6429 | 1.7869 | 0.0339 | 10.1374 | 0 |
| 1.7502 | 0.0561 | 3.2762 | 1.6607 | 0.0360 | 9.6159 | 1 |
| 1.6268 | 0.0595 | 3.0493 | 1.5698 | 0.0374 | 8.6767 | 2 |
| 1.5533 | 0.0614 | 2.3917 | 1.4965 | 0.0387 | 7.2278 | 3 |
| 1.4953 | 0.0631 | 1.8208 | 1.4549 | 0.0397 | 4.4667 | 4 |
| 1.4388 | 0.0647 | 1.7488 | 1.3669 | 0.0411 | 6.2695 | 5 |
| 1.3532 | 0.0670 | 1.6950 | 1.2392 | 0.0438 | 7.1167 | 6 |
| 1.2122 | 0.0710 | 1.6785 | 1.0705 | 0.0471 | 4.7509 | 7 |
| 1.0193 | 0.0768 | 1.8537 | 0.9080 | 0.0502 | 5.7317 | 8 |
| 0.8466 | 0.0823 | 2.3126 | 0.7098 | 0.0543 | 5.7454 | 9 |
| 0.7039 | 0.0870 | 2.3440 | 0.5934 | 0.0567 | 6.9383 | 10 |
| 0.5754 | 0.0914 | 2.3564 | 0.5071 | 0.0586 | 6.6276 | 11 |
| 0.4749 | 0.0949 | 2.6925 | 0.4391 | 0.0601 | 6.8169 | 12 |
| 0.3910 | 0.0978 | 3.0462 | 0.3908 | 0.0612 | 8.1284 | 13 |
| 0.3311 | 0.0999 | 3.3402 | 0.3658 | 0.0617 | 8.6094 | 14 |
Framework versions
- Transformers 4.33.0.dev0
- TensorFlow 2.13.0
- Tokenizers 0.13.3
