metadata
base_model: openai/whisper-small
datasets:
- test16082024Diny
language:
- fr
library_name: peft
license: apache-2.0
tags:
- generated_from_trainer
model-index:
- name: Whisper Small Diny
results: []
Whisper Small Diny
This model is a fine-tuned version of openai/whisper-small on the Diny dataset. It achieves the following results on the evaluation set:
- Loss: 0.6336
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: 0.0001
- train_batch_size: 8
- 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: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.853 | 1.0 | 9 | 3.5040 |
| 3.2074 | 2.0 | 18 | 3.0067 |
| 2.6069 | 3.0 | 27 | 2.4636 |
| 1.955 | 4.0 | 36 | 1.8001 |
| 1.1568 | 5.0 | 45 | 0.9291 |
| 0.7625 | 6.0 | 54 | 0.8403 |
| 0.585 | 7.0 | 63 | 0.7019 |
| 0.5889 | 8.0 | 72 | 0.7064 |
| 0.5227 | 9.0 | 81 | 0.6428 |
| 0.5145 | 10.0 | 90 | 0.6336 |
Framework versions
- PEFT 0.12.0
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1