metadata
license: apache-2.0
library_name: peft
tags:
- generated_from_trainer
base_model: openai/whisper-large-v3
model-index:
- name: whisper-large-v3-MH-fine-tuned
results: []
whisper-large-v3-MH-fine-tuned
This model is a fine-tuned version of openai/whisper-large-v3 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0926
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: 10
- eval_batch_size: 8
- 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: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| No log | 1.0 | 1 | 1.5700 |
| No log | 2.0 | 2 | 1.5532 |
| No log | 3.0 | 3 | 1.5169 |
| No log | 4.0 | 4 | 1.4654 |
| No log | 5.0 | 5 | 1.4050 |
| No log | 6.0 | 6 | 1.3413 |
| No log | 7.0 | 7 | 1.2879 |
| No log | 8.0 | 8 | 1.2420 |
| No log | 9.0 | 9 | 1.1945 |
| No log | 10.0 | 10 | 1.1463 |
| No log | 11.0 | 11 | 1.0969 |
| No log | 12.0 | 12 | 1.0539 |
| No log | 13.0 | 13 | 1.0243 |
| No log | 14.0 | 14 | 1.0018 |
| No log | 15.0 | 15 | 0.9775 |
| No log | 16.0 | 16 | 0.9583 |
| No log | 17.0 | 17 | 0.9521 |
| No log | 18.0 | 18 | 0.9549 |
| No log | 19.0 | 19 | 0.9596 |
| No log | 20.0 | 20 | 0.9569 |
| No log | 21.0 | 21 | 0.9512 |
| No log | 22.0 | 22 | 0.9512 |
| No log | 23.0 | 23 | 0.9577 |
| No log | 24.0 | 24 | 0.9615 |
| 1.0177 | 25.0 | 25 | 0.9770 |
| 1.0177 | 26.0 | 26 | 0.9980 |
| 1.0177 | 27.0 | 27 | 1.0184 |
| 1.0177 | 28.0 | 28 | 1.0271 |
| 1.0177 | 29.0 | 29 | 1.0317 |
| 1.0177 | 30.0 | 30 | 1.0514 |
| 1.0177 | 31.0 | 31 | 1.0524 |
| 1.0177 | 32.0 | 32 | 1.0568 |
| 1.0177 | 33.0 | 33 | 1.0673 |
| 1.0177 | 34.0 | 34 | 1.0699 |
| 1.0177 | 35.0 | 35 | 1.0774 |
| 1.0177 | 36.0 | 36 | 1.0873 |
| 1.0177 | 37.0 | 37 | 1.0838 |
| 1.0177 | 38.0 | 38 | 1.0742 |
| 1.0177 | 39.0 | 39 | 1.0757 |
| 1.0177 | 40.0 | 40 | 1.0692 |
| 1.0177 | 41.0 | 41 | 1.0705 |
| 1.0177 | 42.0 | 42 | 1.0738 |
| 1.0177 | 43.0 | 43 | 1.0765 |
| 1.0177 | 44.0 | 44 | 1.0762 |
| 1.0177 | 45.0 | 45 | 1.0806 |
| 1.0177 | 46.0 | 46 | 1.0755 |
| 1.0177 | 47.0 | 47 | 1.0789 |
| 1.0177 | 48.0 | 48 | 1.0783 |
| 1.0177 | 49.0 | 49 | 1.0856 |
| 0.2796 | 50.0 | 50 | 1.0926 |
Framework versions
- PEFT 0.11.1.dev0
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1