<!-- 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-science-v4-e6-manual
This model is a fine-tuned version of theojolliffe/bart-cnn-science on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 2.0752
- Rouge1: 49.2922
- Rouge2: 27.0916
- Rougel: 29.2754
- Rougelsum: 46.4762
- Gen Len: 140.8
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: 6
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 42 | 1.5703 | 45.3807 | 24.4571 | 27.901 | 42.7866 | 142.0 |
| No log | 2.0 | 84 | 1.5729 | 46.8902 | 24.2952 | 27.8304 | 43.9581 | 142.0 |
| No log | 3.0 | 126 | 1.7025 | 48.836 | 26.917 | 30.1325 | 45.7887 | 142.0 |
| No log | 4.0 | 168 | 1.8526 | 48.906 | 26.8641 | 30.4677 | 46.1825 | 139.2 |
| No log | 5.0 | 210 | 1.9818 | 51.145 | 28.834 | 30.1862 | 48.6876 | 141.8 |
| No log | 6.0 | 252 | 2.0752 | 49.2922 | 27.0916 | 29.2754 | 46.4762 | 140.8 |
Framework versions
- Transformers 4.20.0
- Pytorch 1.11.0+cu113
- Datasets 2.3.2
- Tokenizers 0.12.1
