metadata
tags:
- simplification
- generated_from_trainer
metrics:
- rouge
model-index:
- name: marimari-r2r-mlsum-clara-med
results: []
marimari-r2r-mlsum-clara-med
This model is a fine-tuned version of IIC/marimari-r2r-mlsum on the None dataset. It achieves the following results on the evaluation set:
- Loss: 4.0687
- Rouge1: 41.588
- Rouge2: 23.5096
- Rougel: 35.9281
- Rougelsum: 36.128
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: 5.6e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 190 | 2.3804 | 41.7652 | 24.3329 | 36.7515 | 36.8694 |
| No log | 2.0 | 380 | 2.3417 | 41.586 | 23.679 | 35.8653 | 36.065 |
| 1.9588 | 3.0 | 570 | 2.4777 | 40.954 | 23.0312 | 35.4667 | 35.7321 |
| 1.9588 | 4.0 | 760 | 2.8230 | 41.387 | 23.4641 | 35.9558 | 36.1302 |
| 0.6288 | 5.0 | 950 | 3.0836 | 40.7233 | 22.4476 | 34.852 | 35.1656 |
| 0.6288 | 6.0 | 1140 | 3.2078 | 40.8334 | 22.4327 | 35.0436 | 35.2623 |
| 0.6288 | 7.0 | 1330 | 3.3649 | 40.6737 | 22.4294 | 34.5433 | 34.9343 |
| 0.1473 | 8.0 | 1520 | 3.4503 | 40.8818 | 22.6808 | 34.6777 | 34.9179 |
| 0.1473 | 9.0 | 1710 | 3.5140 | 40.4208 | 22.2582 | 34.5103 | 34.8161 |
| 0.0706 | 10.0 | 1900 | 3.5805 | 40.6348 | 22.4714 | 34.6782 | 34.9531 |
| 0.0706 | 11.0 | 2090 | 3.6325 | 40.932 | 22.4958 | 34.7695 | 35.0314 |
| 0.0706 | 12.0 | 2280 | 3.6405 | 40.619 | 22.406 | 34.9997 | 35.3007 |
| 0.0401 | 13.0 | 2470 | 3.7279 | 40.7365 | 22.2549 | 34.6789 | 34.9794 |
| 0.0401 | 14.0 | 2660 | 3.7440 | 41.1684 | 23.1526 | 35.4117 | 35.7039 |
| 0.0277 | 15.0 | 2850 | 3.8185 | 41.3103 | 23.52 | 35.5945 | 35.9176 |
| 0.0277 | 16.0 | 3040 | 3.8215 | 40.5096 | 22.3435 | 34.7064 | 35.0037 |
| 0.0277 | 17.0 | 3230 | 3.8925 | 41.3644 | 23.2827 | 35.3861 | 35.6649 |
| 0.0172 | 18.0 | 3420 | 3.8594 | 41.6572 | 23.574 | 35.5946 | 35.8383 |
| 0.0172 | 19.0 | 3610 | 3.9191 | 41.4862 | 23.2408 | 35.5638 | 35.7895 |
| 0.0087 | 20.0 | 3800 | 3.8776 | 41.8812 | 23.6052 | 35.9983 | 36.2555 |
| 0.0087 | 21.0 | 3990 | 3.9526 | 42.0435 | 23.6758 | 36.0881 | 36.3736 |
| 0.0087 | 22.0 | 4180 | 3.9847 | 41.7187 | 23.5729 | 36.0514 | 36.3019 |
| 0.0036 | 23.0 | 4370 | 3.9939 | 41.6098 | 23.3451 | 35.779 | 35.9889 |
| 0.0036 | 24.0 | 4560 | 4.0194 | 41.1443 | 23.1271 | 35.6529 | 35.808 |
| 0.002 | 25.0 | 4750 | 4.0231 | 41.5422 | 23.5603 | 35.8412 | 36.0677 |
| 0.002 | 26.0 | 4940 | 4.0439 | 41.5561 | 23.5496 | 35.8154 | 36.0846 |
| 0.002 | 27.0 | 5130 | 4.0554 | 41.6566 | 23.4052 | 35.8392 | 36.0672 |
| 0.0014 | 28.0 | 5320 | 4.0610 | 41.6654 | 23.5138 | 35.9715 | 36.1973 |
| 0.0014 | 29.0 | 5510 | 4.0658 | 41.5467 | 23.464 | 35.7852 | 36.0226 |
| 0.0011 | 30.0 | 5700 | 4.0687 | 41.588 | 23.5096 | 35.9281 | 36.128 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0
- Datasets 2.8.0
- Tokenizers 0.12.1