<!-- 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. -->
distilbart-cnn-arxiv-pubmed-v3-e12
This model is a fine-tuned version of theojolliffe/distilbart-cnn-arxiv-pubmed on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8157
- Rouge1: 56.7429
- Rouge2: 41.0185
- Rougel: 44.1014
- Rougelsum: 54.8121
- Gen Len: 142.0
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 12
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
---|---|---|---|---|---|---|---|---|
1.5037 | 1.0 | 795 | 1.0815 | 52.4727 | 33.4915 | 35.3774 | 50.1955 | 142.0 |
0.8894 | 2.0 | 1590 | 0.9462 | 52.8867 | 34.0406 | 36.5249 | 50.4636 | 141.5741 |
0.7037 | 3.0 | 2385 | 0.8841 | 53.7966 | 35.0969 | 38.4158 | 51.3369 | 142.0 |
0.4914 | 4.0 | 3180 | 0.8437 | 52.6766 | 34.0573 | 36.8907 | 50.3088 | 142.0 |
0.3945 | 5.0 | 3975 | 0.8067 | 54.3147 | 36.2081 | 39.6366 | 52.1494 | 142.0 |
0.2799 | 6.0 | 4770 | 0.8403 | 54.2813 | 37.0786 | 39.9196 | 51.9176 | 141.9815 |
0.2211 | 7.0 | 5565 | 0.8207 | 53.9403 | 36.517 | 39.0372 | 51.4491 | 141.9815 |
0.1795 | 8.0 | 6360 | 0.8014 | 55.6607 | 39.3082 | 41.8295 | 53.4674 | 142.0 |
0.1428 | 9.0 | 7155 | 0.8051 | 55.0575 | 38.823 | 41.8849 | 52.9606 | 142.0 |
0.1358 | 10.0 | 7950 | 0.8149 | 56.6986 | 41.0 | 43.5207 | 54.6402 | 142.0 |
0.1122 | 11.0 | 8745 | 0.8134 | 56.5416 | 40.9495 | 44.2989 | 54.5623 | 142.0 |
0.0873 | 12.0 | 9540 | 0.8157 | 56.7429 | 41.0185 | 44.1014 | 54.8121 | 142.0 |
Framework versions
- Transformers 4.18.0
- Pytorch 1.11.0+cu113
- Datasets 2.1.0
- Tokenizers 0.12.1