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: 0.9861
- Rouge1: 53.286
- Rouge2: 27.5382
- Rougel: 41.0668
- Rougelsum: 41.0378
- Gen Len: 43.05
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: 5e-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: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 2.8241 | 1.0 | 983 | 2.1249 | 38.9009 | 11.883 | 30.4391 | 30.3769 | 40.86 |
| 2.2297 | 2.0 | 1966 | 1.8067 | 40.8363 | 15.1119 | 32.2579 | 32.2289 | 38.59 |
| 1.9156 | 3.0 | 2949 | 1.4818 | 43.5357 | 16.8437 | 33.6575 | 33.6616 | 35.11 |
| 1.5954 | 4.0 | 3932 | 1.1631 | 48.9965 | 22.296 | 36.029 | 35.984 | 41.38 |
| 1.2973 | 5.0 | 4915 | 0.9861 | 53.286 | 27.5382 | 41.0668 | 41.0378 | 43.05 |
Framework versions
- Transformers 4.31.0
- Pytorch 2.0.1+cu117
- Datasets 2.14.2
- Tokenizers 0.13.3