<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
tamasheq-98-1
This model is a fine-tuned version of ad019el/tamasheq-98 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.6353
- Wer: 0.9704
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 200
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Wer |
|---|---|---|---|---|
| 3.0938 | 10.0 | 500 | 1.8353 | 1.0 |
| 0.3229 | 20.0 | 1000 | 1.6417 | 0.9815 |
| 0.224 | 30.0 | 1500 | 1.5298 | 0.9593 |
| 0.1906 | 40.0 | 2000 | 1.5385 | 0.9407 |
| 0.1692 | 50.0 | 2500 | 1.5741 | 1.0519 |
| 0.1581 | 60.0 | 3000 | 1.6353 | 0.9704 |
Framework versions
- Transformers 4.20.1
- Pytorch 1.11.0
- Datasets 2.1.0
- Tokenizers 0.12.1
