xsum-summarization-t5-small
This model is a fine-tuned version of t5-small on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.8569
- Rouge1: 0.2359
- Rouge2: 0.0473
- Rougel: 0.1831
- Rougelsum: 0.1832
- Meteor: 0.1465
- Bertscore: 0.192
- Gen Len: 19.516
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: 5e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Use 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: 500
- num_epochs: 7
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Meteor | Bertscore | Gen Len |
|---|---|---|---|---|---|---|---|---|---|---|
| 3.8387 | 1.0 | 625 | 3.2061 | 0.1833 | 0.0287 | 0.1445 | 0.1445 | 0.1098 | 0.1257 | 19.8657 |
| 3.297 | 2.0 | 1250 | 2.9853 | 0.2064 | 0.0379 | 0.1618 | 0.1618 | 0.1256 | 0.154 | 19.4563 |
| 3.1871 | 3.0 | 1875 | 2.9210 | 0.2191 | 0.0412 | 0.1714 | 0.1714 | 0.1341 | 0.1731 | 19.3843 |
| 3.1439 | 4.0 | 2500 | 2.8890 | 0.2275 | 0.0448 | 0.1772 | 0.1772 | 0.1402 | 0.1815 | 19.4417 |
| 3.1128 | 5.0 | 3125 | 2.8696 | 0.2324 | 0.0461 | 0.1809 | 0.1809 | 0.1437 | 0.1873 | 19.49 |
| 3.0986 | 6.0 | 3750 | 2.8598 | 0.2353 | 0.0469 | 0.1828 | 0.1829 | 0.146 | 0.1909 | 19.5097 |
| 3.0899 | 7.0 | 4375 | 2.8569 | 0.2359 | 0.0473 | 0.1831 | 0.1832 | 0.1465 | 0.192 | 19.516 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.5.1+cu121
- Datasets 3.5.1
- Tokenizers 0.21.1