mms-300m-nyagen-original-train-ft-62
This model is a fine-tuned version of facebook/mms-300m on the NYAGEN - NA dataset. It achieves the following results on the evaluation set:
- Loss: 0.4169
- Wer: 0.3193
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 16
- 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: 30.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 0.4974 | 0.9940 | 500 | 0.6994 | 0.6079 |
| 0.2261 | 1.9881 | 1000 | 0.4816 | 0.3740 |
| 0.1722 | 2.9821 | 1500 | 0.4425 | 0.3443 |
| 0.1615 | 3.9761 | 2000 | 0.4169 | 0.3193 |
| 0.112 | 4.9702 | 2500 | 0.4534 | 0.3031 |
| 0.1122 | 5.9642 | 3000 | 0.4591 | 0.2974 |
| 0.0865 | 6.9583 | 3500 | 0.4723 | 0.2928 |
| 0.0763 | 7.9523 | 4000 | 0.4821 | 0.2883 |
| 0.0622 | 8.9463 | 4500 | 0.5079 | 0.2855 |
Framework versions
- Transformers 4.53.0
- Pytorch 2.10.0+cu128
- Datasets 4.5.0
- Tokenizers 0.21.4