<!-- 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-model2-1510-e8
This model is a fine-tuned version of theojolliffe/bart-paraphrase-v4-e1-feedback on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3655
- Rouge1: 61.3129
- Rouge2: 57.3305
- Rougel: 60.8028
- Rougelsum: 60.5111
- Gen Len: 20.0
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: 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 | 409 | 0.4572 | 56.7459 | 47.5708 | 54.6144 | 54.9188 | 20.0 |
| 0.5704 | 2.0 | 818 | 0.4349 | 58.4751 | 50.7958 | 56.5975 | 56.941 | 20.0 |
| 0.1956 | 3.0 | 1227 | 0.3952 | 61.6499 | 55.4368 | 60.157 | 60.2961 | 20.0 |
| 0.1177 | 4.0 | 1636 | 0.3685 | 59.8851 | 54.1843 | 58.6443 | 58.8519 | 20.0 |
| 0.0752 | 5.0 | 2045 | 0.3654 | 60.975 | 55.9124 | 60.0336 | 59.8978 | 20.0 |
| 0.0752 | 6.0 | 2454 | 0.3525 | 61.268 | 55.7247 | 60.2274 | 60.1515 | 20.0 |
| 0.0526 | 7.0 | 2863 | 0.3519 | 61.6626 | 57.9242 | 61.0212 | 60.8486 | 20.0 |
| 0.0388 | 8.0 | 3272 | 0.3655 | 61.3129 | 57.3305 | 60.8028 | 60.5111 | 20.0 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1
