metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: openai/whisper-small
model-index:
- name: Whisper-small-speechocean
results: []
Whisper-small-speechocean
This model is a fine-tuned version of openai/whisper-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.5759
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:
- learning_rate: 0.001
- train_batch_size: 6
- eval_batch_size: 6
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.9894 | 1.0 | 417 | 1.7665 |
| 0.8028 | 2.0 | 834 | 0.7597 |
| 0.6486 | 3.0 | 1251 | 0.5817 |
| 0.471 | 4.0 | 1668 | 0.5405 |
| 0.3168 | 5.0 | 2085 | 0.5099 |
| 0.2274 | 6.0 | 2502 | 0.5287 |
| 0.1328 | 7.0 | 2919 | 0.5458 |
| 0.0787 | 8.0 | 3336 | 0.5511 |
| 0.0306 | 9.0 | 3753 | 0.5689 |
| 0.0179 | 10.0 | 4170 | 0.5759 |
Framework versions
- PEFT 0.8.0
- Transformers 4.40.0
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1