wav2vec2-nigerian-lid-v3
This model is a fine-tuned version of facebook/wav2vec2-large-xlsr-53 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1276
- Accuracy: 0.9791
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: 0.0002
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.1
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.3345 | 1.0 | 280 | 0.3439 | 0.9280 |
| 0.0597 | 2.0 | 560 | 0.1034 | 0.9838 |
| 0.0323 | 3.0 | 840 | 0.1233 | 0.9781 |
| 0.0322 | 4.0 | 1120 | 0.1209 | 0.9776 |
| 0.0254 | 5.0 | 1400 | 0.1961 | 0.9755 |
| 0.0103 | 6.0 | 1680 | 0.1063 | 0.9807 |
| 0.0116 | 7.0 | 1960 | 0.0669 | 0.9880 |
| 0.0003 | 8.0 | 2240 | 0.1446 | 0.9791 |
| 0.0026 | 9.0 | 2520 | 0.1233 | 0.9802 |
| 0.0021 | 10.0 | 2800 | 0.1276 | 0.9791 |
Framework versions
- Transformers 5.0.0
- Pytorch 2.10.0+cu128
- Datasets 4.0.0
- Tokenizers 0.22.2
- Downloads last month
- 32