<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
ainize-kobart-news-eb-finetuned-papers
This model is a fine-tuned version of ainize/kobart-news on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3066
- Rouge1: 14.5433
- Rouge2: 5.2238
- Rougel: 14.4731
- Rougelsum: 14.5183
- Gen Len: 19.9934
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: 4
- eval_batch_size: 4
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 0.1918 | 1.0 | 7200 | 0.2403 | 14.6883 | 5.2427 | 14.6306 | 14.6489 | 19.9938 |
| 0.1332 | 2.0 | 14400 | 0.2391 | 14.5165 | 5.2443 | 14.493 | 14.4908 | 19.9972 |
| 0.0966 | 3.0 | 21600 | 0.2539 | 14.758 | 5.4976 | 14.6906 | 14.7188 | 19.9941 |
| 0.0736 | 4.0 | 28800 | 0.2782 | 14.6267 | 5.3371 | 14.5578 | 14.6014 | 19.9934 |
| 0.0547 | 5.0 | 36000 | 0.3066 | 14.5433 | 5.2238 | 14.4731 | 14.5183 | 19.9934 |
Framework versions
- Transformers 4.19.4
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
