<!-- 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. -->
pegasus-newsroom-cnn1_50k
This model is a fine-tuned version of oMateos2020/pegasus-newsroom-cnn1_50k on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.1267
- Rouge1: 38.0081
- Rouge2: 16.5536
- Rougel: 26.4916
- Rougelsum: 35.1349
- Gen Len: 59.4912
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: 0.0001
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 32
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 500
- num_epochs: 5
- mixed_precision_training: Native AMP
- label_smoothing_factor: 0.1
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum | Gen Len |
|---|---|---|---|---|---|---|---|---|
| 3.144 | 0.26 | 100 | 3.0323 | 38.3168 | 16.7528 | 26.2646 | 35.2447 | 66.2372 |
| 3.0556 | 0.51 | 200 | 3.0351 | 38.39 | 16.8027 | 26.3412 | 35.37 | 67.4676 |
| 3.0701 | 0.77 | 300 | 3.0345 | 38.5742 | 16.922 | 26.3568 | 35.51 | 68.662 |
| 3.1679 | 1.03 | 400 | 3.0321 | 38.5319 | 16.8049 | 26.4933 | 35.4775 | 65.976 |
| 3.1041 | 1.28 | 500 | 3.0246 | 38.1381 | 16.63 | 26.2484 | 35.0999 | 64.6896 |
| 3.0352 | 1.54 | 600 | 3.0206 | 38.9063 | 17.0281 | 27.0288 | 35.9175 | 59.0668 |
| 3.0894 | 1.79 | 700 | 3.0251 | 38.4461 | 16.7732 | 26.4394 | 35.4807 | 63.2792 |
| 3.0529 | 2.05 | 800 | 3.0400 | 38.5088 | 16.8921 | 26.5526 | 35.5236 | 64.294 |
| 3.0002 | 2.31 | 900 | 3.0394 | 38.6899 | 16.8703 | 26.6771 | 35.6207 | 62.8004 |
| 3.0167 | 2.56 | 1000 | 3.0394 | 38.3532 | 16.6176 | 26.5433 | 35.3282 | 61.63 |
| 3.0168 | 2.82 | 1100 | 3.0421 | 38.7613 | 17.0107 | 26.8424 | 35.7508 | 62.67 |
| 3.0412 | 3.08 | 1200 | 3.0491 | 38.6132 | 16.8046 | 26.61 | 35.6002 | 61.7924 |
| 3.1273 | 3.33 | 1300 | 3.0823 | 38.5498 | 16.795 | 26.5569 | 35.613 | 60.6872 |
| 3.0634 | 3.59 | 1400 | 3.1010 | 38.0927 | 16.4367 | 26.2315 | 35.1311 | 59.252 |
| 3.097 | 3.84 | 1500 | 3.1147 | 37.7644 | 16.3156 | 26.2674 | 34.8315 | 59.7592 |
| 3.1287 | 4.1 | 1600 | 3.1267 | 38.0081 | 16.5536 | 26.4916 | 35.1349 | 59.4912 |
Framework versions
- Transformers 4.21.0
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1
