metadata
license: apache-2.0
base_model: openai/whisper-base.en
tags:
- generated_from_trainer
metrics:
- rouge
model-index:
- name: whispherMusic
results: []
whispherMusic
This model is a fine-tuned version of openai/whisper-base.en on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4596
- Rouge1: 67.0305
- Rouge2: 49.4145
- Rougel: 59.4632
- Rougelsum: 59.5132
- Gen Len: 47.98
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.5281 | 1.0 | 983 | 1.7656 | 47.2738 | 19.609 | 35.166 | 35.1418 | 39.81 |
| 1.8844 | 2.0 | 1966 | 1.4416 | 48.4369 | 22.034 | 36.3312 | 36.3541 | 41.88 |
| 1.6071 | 3.0 | 2949 | 1.2060 | 49.7668 | 24.7504 | 38.8626 | 38.8523 | 37.8 |
| 1.3867 | 4.0 | 3932 | 1.0056 | 52.8821 | 28.115 | 40.5389 | 40.5263 | 44.19 |
| 1.1964 | 5.0 | 4915 | 0.8299 | 56.3235 | 32.1936 | 44.4629 | 44.3334 | 44.24 |
| 1.0301 | 6.0 | 5898 | 0.6930 | 59.6505 | 36.3548 | 48.4558 | 48.3634 | 44.14 |
| 0.904 | 7.0 | 6881 | 0.5916 | 61.3377 | 40.3586 | 50.8339 | 50.7188 | 46.15 |
| 0.7985 | 8.0 | 7864 | 0.5157 | 64.7817 | 45.3792 | 56.0262 | 55.9178 | 47.7 |
| 0.7125 | 9.0 | 8847 | 0.4722 | 66.0546 | 47.0456 | 57.422 | 57.37 | 48.35 |
| 0.6514 | 10.0 | 9830 | 0.4596 | 67.0305 | 49.4145 | 59.4632 | 59.5132 | 47.98 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.2
- Tokenizers 0.13.3