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: 0.6849
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: 4
- 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 | 1.0192 |
| No log | 2.0 | 2 | 1.0085 |
| No log | 3.0 | 3 | 0.9859 |
| No log | 4.0 | 4 | 0.9554 |
| No log | 5.0 | 5 | 0.9172 |
| No log | 6.0 | 6 | 0.8728 |
| No log | 7.0 | 7 | 0.8330 |
| No log | 8.0 | 8 | 0.7908 |
| No log | 9.0 | 9 | 0.7546 |
| No log | 10.0 | 10 | 0.7167 |
| No log | 11.0 | 11 | 0.6891 |
| No log | 12.0 | 12 | 0.6698 |
| No log | 13.0 | 13 | 0.6546 |
| No log | 14.0 | 14 | 0.6372 |
| No log | 15.0 | 15 | 0.6247 |
| No log | 16.0 | 16 | 0.6198 |
| No log | 17.0 | 17 | 0.6209 |
| No log | 18.0 | 18 | 0.6249 |
| No log | 19.0 | 19 | 0.6316 |
| No log | 20.0 | 20 | 0.6368 |
| No log | 21.0 | 21 | 0.6398 |
| No log | 22.0 | 22 | 0.6463 |
| No log | 23.0 | 23 | 0.6555 |
| No log | 24.0 | 24 | 0.6680 |
| 0.4535 | 25.0 | 25 | 0.6789 |
| 0.4535 | 26.0 | 26 | 0.6905 |
| 0.4535 | 27.0 | 27 | 0.7006 |
| 0.4535 | 28.0 | 28 | 0.7102 |
| 0.4535 | 29.0 | 29 | 0.7208 |
| 0.4535 | 30.0 | 30 | 0.7363 |
| 0.4535 | 31.0 | 31 | 0.7512 |
| 0.4535 | 32.0 | 32 | 0.7582 |
| 0.4535 | 33.0 | 33 | 0.7685 |
| 0.4535 | 34.0 | 34 | 0.7794 |
| 0.4535 | 35.0 | 35 | 0.7849 |
| 0.4535 | 36.0 | 36 | 0.7896 |
| 0.4535 | 37.0 | 37 | 0.7917 |
| 0.4535 | 38.0 | 38 | 0.8132 |
| 0.4535 | 39.0 | 39 | 0.7889 |
| 0.4535 | 40.0 | 40 | 0.7614 |
| 0.4535 | 41.0 | 41 | 0.7371 |
| 0.4535 | 42.0 | 42 | 0.7206 |
| 0.4535 | 43.0 | 43 | 0.7066 |
| 0.4535 | 44.0 | 44 | 0.7024 |
| 0.4535 | 45.0 | 45 | 0.7128 |
| 0.4535 | 46.0 | 46 | 0.7242 |
| 0.4535 | 47.0 | 47 | 0.7137 |
| 0.4535 | 48.0 | 48 | 0.7004 |
| 0.4535 | 49.0 | 49 | 0.6883 |
| 0.0261 | 50.0 | 50 | 0.6849 |
Framework versions
- PEFT 0.11.2.dev0
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1