flan-t5-small-finetuned-scientific-articles
This model is a fine-tuned version of google/flan-t5-small on the scientific_papers dataset. It achieves the following results on the evaluation set:
- Loss: 2.6792
- Rouge1: 8.0297
- Rouge2: 2.5421
- Rougel: 6.6908
- Rougelsum: 7.3431
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: 5.6e-05
- train_batch_size: 9
- eval_batch_size: 9
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| 3.0943 | 1.0 | 56 | 2.8262 | 3.9456 | 1.1211 | 3.3527 | 3.6682 |
| 3.0216 | 2.0 | 112 | 2.7682 | 6.0659 | 1.6822 | 5.2499 | 5.7102 |
| 2.9495 | 3.0 | 168 | 2.7316 | 7.4704 | 2.4232 | 6.2443 | 6.913 |
| 2.9057 | 4.0 | 224 | 2.7050 | 7.8789 | 2.6559 | 6.5559 | 7.1858 |
| 2.8622 | 5.0 | 280 | 2.6792 | 8.0297 | 2.5421 | 6.6908 | 7.3431 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1