metadata
library_name: transformers
license: cc-by-nc-4.0
base_model: facebook/mms-1b-all
tags:
- generated_from_trainer
metrics:
- wer
model-index:
- name: wav2vec2-large-mms-1b-dzo_M3
results: []
wav2vec2-large-mms-1b-dzo_M3
This model is a fine-tuned version of facebook/mms-1b-all on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2959
- Wer: 0.3420
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: 0.004
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 1.1643 | 1.0 | 25 | 0.7774 | 0.5192 |
| 0.7917 | 2.0 | 50 | 0.5777 | 0.4425 |
| 0.7344 | 3.0 | 75 | 0.4905 | 0.4186 |
| 0.6428 | 4.0 | 100 | 0.4362 | 0.4003 |
| 0.5454 | 5.0 | 125 | 0.3963 | 0.3855 |
| 0.5281 | 6.0 | 150 | 0.3619 | 0.3718 |
| 0.5125 | 7.0 | 175 | 0.3410 | 0.3623 |
| 0.4359 | 8.0 | 200 | 0.3188 | 0.3529 |
| 0.4189 | 9.0 | 225 | 0.3006 | 0.3394 |
| 0.3566 | 10.0 | 250 | 0.2959 | 0.3420 |
Framework versions
- Transformers 4.49.0.dev0
- Pytorch 2.5.1+cu121
- Datasets 3.2.0
- Tokenizers 0.21.0