TUS-swahili-news
This model is a fine-tuned version of eolang/TUS on the swahili_news dataset. It achieves the following results on the evaluation set:
- Loss: 6.6283
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: 2e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 7.1478 | 1.0 | 1010 | 6.8678 |
| 6.8352 | 2.0 | 2020 | 6.6895 |
| 6.7224 | 3.0 | 3030 | 6.6197 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.1
- Datasets 2.8.0
- Tokenizers 0.13.2
- Downloads last month
- 6