metadata
license: apache-2.0
base_model: openai/whisper-medium
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: whisper-med-LoRA_FAA_data_small
results: []
whisper-med-LoRA_FAA_data_small
This model is a fine-tuned version of openai/whisper-medium on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 4.1617
- Wer: 56.9444
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: 16
- eval_batch_size: 16
- 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 | Wer |
|---|---|---|---|---|
| 4.5172 | 1.0 | 2 | 4.4887 | 56.9444 |
| 4.3753 | 2.0 | 4 | 4.4566 | 56.9444 |
| 4.5353 | 3.0 | 6 | 4.3517 | 56.9444 |
| 4.3463 | 4.0 | 8 | 4.2345 | 56.9444 |
| 4.1617 | 5.0 | 10 | 4.2299 | 56.9444 |
| 4.3382 | 6.0 | 12 | 4.2150 | 56.9444 |
| 4.2763 | 7.0 | 14 | 4.1884 | 56.9444 |
| 4.2853 | 8.0 | 16 | 4.1740 | 56.9444 |
| 4.2673 | 9.0 | 18 | 4.1669 | 56.9444 |
| 4.1283 | 10.0 | 20 | 4.1617 | 56.9444 |
Framework versions
- Transformers 4.33.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.5
- Tokenizers 0.13.3