<!-- 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. -->
bart-cnn-pubmed-arxiv-pubmed-v3-e8
This model is a fine-tuned version of theojolliffe/bart-cnn-pubmed-arxiv-pubmed on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.7778
- Rouge1: 55.6307
- Rouge2: 38.1306
- Rougel: 40.7127
- Rougelsum: 53.3739
- Gen Len: 141.9815
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: 2
- eval_batch_size: 2
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 398 | 0.9563 | 53.0477 | 33.0365 | 35.4483 | 50.5525 | 142.0 |
| 1.1233 | 2.0 | 796 | 0.8260 | 53.8629 | 34.5031 | 37.08 | 51.129 | 142.0 |
| 0.6753 | 3.0 | 1194 | 0.7898 | 53.6508 | 34.7559 | 37.0541 | 50.7535 | 142.0 |
| 0.4532 | 4.0 | 1592 | 0.7765 | 53.2109 | 34.5657 | 37.3743 | 50.9145 | 142.0 |
| 0.4532 | 5.0 | 1990 | 0.7551 | 55.0766 | 37.5722 | 40.0653 | 52.5655 | 142.0 |
| 0.3142 | 6.0 | 2388 | 0.7744 | 54.7674 | 36.7664 | 39.9027 | 52.1542 | 142.0 |
| 0.2257 | 7.0 | 2786 | 0.7728 | 55.6258 | 37.9929 | 40.8985 | 53.4423 | 142.0 |
| 0.1674 | 8.0 | 3184 | 0.7778 | 55.6307 | 38.1306 | 40.7127 | 53.3739 | 141.9815 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1
