Configuration Parsing
Warning:
In adapter_config.json: "peft.task_type" must be a string
Whisper Turbo ko
This model is a fine-tuned version of openai/whisper-large-v3-turbo on the custom dataset. It achieves the following results on the evaluation set:
- Loss: 0.0965
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: 64
- eval_batch_size: 256
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 200
- training_steps: 1000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.3036 | 0.5435 | 100 | 0.6036 |
| 0.1076 | 1.0870 | 200 | 0.3805 |
| 0.085 | 1.6304 | 300 | 0.2928 |
| 0.0776 | 2.1739 | 400 | 0.2214 |
| 0.0587 | 2.7174 | 500 | 0.1791 |
| 0.0527 | 3.2609 | 600 | 0.1824 |
| 0.04 | 3.8043 | 700 | 0.1482 |
| 0.0295 | 4.3478 | 800 | 0.1285 |
| 0.0261 | 4.8913 | 900 | 0.1167 |
| 0.0297 | 5.4348 | 1000 | 0.0965 |
Framework versions
- PEFT 0.14.0
- Transformers 4.47.1
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0
- Downloads last month
- 2