summarise_v5
This model is a fine-tuned version of allenai/led-base-16384 on the multi_news dataset. It achieves the following results on the evaluation set:
- Loss: 2.3252
- Rouge2 Precision: 0.1458
- Rouge2 Recall: 0.1306
- Rouge2 Fmeasure: 0.1343
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: 5e-05
- train_batch_size: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge2 Precision | Rouge2 Recall | Rouge2 Fmeasure |
|---|---|---|---|---|---|---|
| 2.6266 | 0.13 | 10 | 2.4604 | 0.1021 | 0.179 | 0.124 |
| 2.4818 | 0.27 | 20 | 2.4122 | 0.1402 | 0.1422 | 0.1345 |
| 2.3451 | 0.4 | 30 | 2.3846 | 0.1631 | 0.1177 | 0.1307 |
| 2.4462 | 0.53 | 40 | 2.3584 | 0.1671 | 0.1175 | 0.133 |
| 2.443 | 0.67 | 50 | 2.3395 | 0.1444 | 0.1359 | 0.1344 |
| 2.3822 | 0.8 | 60 | 2.3377 | 0.1517 | 0.1411 | 0.1395 |
| 2.4304 | 0.93 | 70 | 2.3252 | 0.1458 | 0.1306 | 0.1343 |
Framework versions
- Transformers 4.21.3
- Pytorch 1.12.1+cu113
- Datasets 2.6.2.dev0
- Tokenizers 0.12.1