whisper-large-v2_lora_int8
This model is a fine-tuned version of openai/whisper-large-v2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1738
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: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- training_steps: 5000
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.7199 | 0.0 | 1 | 2.3895 |
| 0.1977 | 0.29 | 1000 | 0.1973 |
| 0.1777 | 0.58 | 2000 | 0.1846 |
| 0.1693 | 0.88 | 3000 | 0.1780 |
| 0.1579 | 1.17 | 4000 | 0.1753 |
| 0.157 | 1.46 | 5000 | 0.1738 |
Framework versions
- Transformers 4.27.4
- Pytorch 2.0.0
- Datasets 2.11.0
- Tokenizers 0.13.3