T5-Small-Sinhala-Sumarization-test3
This model is a fine-tuned version of Malmika/T5-Small-Sinhala-Sumarization on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0304
- Rouge1: 0.1355
- Rouge2: 0.0618
- Rougel: 0.1354
- Rougelsum: 0.1356
- Gen Len: 17.8198
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: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.0959 | 1.0 | 4333 | 0.0560 | 0.1357 | 0.062 | 0.1357 | 0.1358 | 17.8575 |
| 0.0531 | 2.0 | 8666 | 0.0367 | 0.1355 | 0.0619 | 0.1355 | 0.1357 | 17.8214 |
| 0.0406 | 3.0 | 12999 | 0.0350 | 0.1355 | 0.0619 | 0.1355 | 0.1357 | 17.8213 |
| 0.0342 | 4.0 | 17332 | 0.0328 | 0.1355 | 0.0618 | 0.1354 | 0.1356 | 17.8198 |
| 0.0323 | 5.0 | 21665 | 0.0304 | 0.1355 | 0.0618 | 0.1354 | 0.1356 | 17.8198 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.1.0+cu121
- Datasets 2.17.0
- Tokenizers 0.15.1