xlmr-singlish-sinhala-lora
This model is a fine-tuned version of deshanksuman/swaBhahsa_XLMR_Transliteration on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1686
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- 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
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.8855 | 1.0 | 313 | 0.1984 |
| 0.7692 | 2.0 | 626 | 0.1824 |
| 0.6254 | 3.0 | 939 | 0.1742 |
| 0.6276 | 4.0 | 1252 | 0.1704 |
| 0.5605 | 5.0 | 1565 | 0.1686 |
Framework versions
- PEFT 0.18.1
- Transformers 5.1.0
- Pytorch 2.7.1+cu118
- Datasets 4.5.0
- Tokenizers 0.22.2
- Downloads last month
- 1
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for RukDias/xlmr-singlish-sinhala-lora
Base model
FacebookAI/xlm-roberta-base