Whisper openai-whisper-medium-LoRA16-es_ecu911
This model is a fine-tuned version of openai/whisper-medium on the llamadas ecu9111 segmentos dmarquez dataset. It achieves the following results on the evaluation set:
- Loss: 1.5185
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 2
- 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: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.5636 | 1.0 | 53 | 1.7149 |
| 1.4208 | 2.0 | 106 | 1.6081 |
| 1.5851 | 3.0 | 159 | 1.5621 |
| 1.424 | 4.0 | 212 | 1.5265 |
| 1.6457 | 5.0 | 265 | 1.5185 |
Framework versions
- PEFT 0.13.2
- Transformers 4.45.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.20.1