<!-- 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. -->
whisper_4_with_init_sun_syl_wd_0_lr_3en4_0005
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: 4.5246
- Train Accuracy: 0.0119
- Train Wermet: 0.9086
- Train Wermet Syl: 1.0238
- Validation Loss: 3.8947
- Validation Accuracy: 0.0113
- Validation Wermet: 0.9346
- Validation Wermet Syl: 0.8618
- Epoch: 4
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': 0.0003, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False, 'weight_decay_rate': 0}
- training_precision: float32
Training results
| Train Loss | Train Accuracy | Train Wermet | Train Wermet Syl | Validation Loss | Validation Accuracy | Validation Wermet | Validation Wermet Syl | Epoch |
|---|---|---|---|---|---|---|---|---|
| 4.9621 | 0.0111 | 1.3737 | 1.2531 | 3.9542 | 0.0114 | 0.9773 | 0.9699 | 0 |
| 4.6813 | 0.0116 | 0.9321 | 0.9542 | 3.9070 | 0.0114 | 0.9669 | 0.9552 | 1 |
| 4.6267 | 0.0117 | 0.9053 | 0.9424 | 3.9109 | 0.0114 | 0.9465 | 0.9128 | 2 |
| 4.5866 | 0.0118 | 0.8961 | 0.9453 | 3.8787 | 0.0115 | 0.9210 | 0.8988 | 3 |
| 4.5246 | 0.0119 | 0.9086 | 1.0238 | 3.8947 | 0.0113 | 0.9346 | 0.8618 | 4 |
Framework versions
- Transformers 4.34.0.dev0
- TensorFlow 2.13.0
- Tokenizers 0.13.3
