AfriHuBERT-nyagen-female-42
This model is a fine-tuned version of ajesujoba/AfriHuBERT on the NYAGEN - NYA dataset. It achieves the following results on the evaluation set:
- Loss: 2.8695
- Wer: 1.0
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: 0.001
- 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: 100
- num_epochs: 30.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Wer |
---|---|---|---|---|
5.5512 | 1.1117 | 100 | 2.8784 | 1.0 |
2.8606 | 2.2235 | 200 | 2.8762 | 1.0 |
2.8621 | 3.3352 | 300 | 2.8734 | 1.0 |
2.8607 | 4.4469 | 400 | 2.8671 | 1.0 |
2.8577 | 5.5587 | 500 | 2.8753 | 1.0 |
2.8559 | 6.6704 | 600 | 2.8659 | 1.0 |
2.8618 | 7.7821 | 700 | 2.8682 | 1.0 |
2.8547 | 8.8939 | 800 | 2.8689 | 1.0 |
2.8429 | 10.0 | 900 | 2.8722 | 1.0 |
2.8572 | 11.1117 | 1000 | 2.8646 | 1.0 |
2.8581 | 12.2235 | 1100 | 2.8732 | 1.0 |
2.8604 | 13.3352 | 1200 | 2.8693 | 1.0 |
2.8567 | 14.4469 | 1300 | 2.8832 | 1.0 |
2.8592 | 15.5587 | 1400 | 2.8699 | 1.0 |
2.8602 | 16.6704 | 1500 | 2.8732 | 1.0 |
2.8628 | 17.7821 | 1600 | 2.8689 | 1.0 |
2.8623 | 18.8939 | 1700 | 2.8656 | 1.0 |
2.849 | 20.0 | 1800 | 2.8894 | 1.0 |
2.861 | 21.1117 | 1900 | 2.8637 | 1.0 |
2.8586 | 22.2235 | 2000 | 2.8661 | 1.0 |
2.8578 | 23.3352 | 2100 | 2.8656 | 1.0 |
2.8571 | 24.4469 | 2200 | 2.8626 | 1.0 |
2.8551 | 25.5587 | 2300 | 2.8677 | 1.0 |
2.8565 | 26.6704 | 2400 | 2.8655 | 1.0 |
2.8537 | 27.7821 | 2500 | 2.8667 | 1.0 |
2.8542 | 28.8939 | 2600 | 2.8642 | 1.0 |
2.8391 | 30.0 | 2700 | 2.8650 | 1.0 |
Framework versions
- Transformers 4.53.0.dev0
- Pytorch 2.6.0+cu124
- Datasets 3.6.0
- Tokenizers 0.21.0
- Downloads last month
- 4
Inference Providers
NEW
This model isn't deployed by any Inference Provider.
๐
Ask for provider support