joheras's picture
update model card README.md
8115550
metadata
license: cc-by-nc-sa-4.0
tags:
  - simplification
  - generated_from_trainer
metrics:
  - rouge
model-index:
  - name: mt5-simplification-spanish-clara-med
    results: []

mt5-simplification-spanish-clara-med

This model is a fine-tuned version of oskrmiguel/mt5-simplification-spanish on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.9610
  • Rouge1: 33.7922
  • Rouge2: 19.5758
  • Rougel: 31.3737
  • Rougelsum: 31.3428

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 2.6876 32.236 18.2352 29.7852 29.7539
No log 2.0 380 2.4617 32.8521 18.9712 30.4958 30.4635
3.3018 3.0 570 2.3487 33.2554 19.3441 30.9036 30.8525
3.3018 4.0 760 2.2711 33.0105 19.01 30.6851 30.5767
2.7431 5.0 950 2.2254 33.1301 18.9618 30.6744 30.6284
2.7431 6.0 1140 2.1847 33.3701 19.1884 30.9138 30.8611
2.7431 7.0 1330 2.1443 33.3158 19.101 30.8317 30.7747
2.5154 8.0 1520 2.1072 33.1638 19.0139 30.7295 30.7162
2.5154 9.0 1710 2.0989 33.4925 19.2107 31.0253 30.9908
2.3763 10.0 1900 2.0709 33.3007 18.9519 30.847 30.8018
2.3763 11.0 2090 2.0631 33.4689 19.1995 31.0712 31.0327
2.3763 12.0 2280 2.0418 33.2536 19.027 30.898 30.8695
2.2811 13.0 2470 2.0345 33.5097 19.2219 31.1057 31.0683
2.2811 14.0 2660 2.0185 33.3544 19.1241 30.913 30.8873
2.2173 15.0 2850 2.0138 33.3856 19.2065 31.0173 30.9447
2.2173 16.0 3040 2.0019 33.4035 19.1803 31.0154 30.981
2.2173 17.0 3230 1.9977 33.4059 19.3078 31.1196 31.0692
2.1612 18.0 3420 1.9883 33.5097 19.3637 31.0966 31.0554
2.1612 19.0 3610 1.9828 33.4965 19.2754 31.1267 31.1021
2.1115 20.0 3800 1.9834 33.7514 19.5325 31.2833 31.2418
2.1115 21.0 3990 1.9754 33.6193 19.429 31.2721 31.2267
2.1115 22.0 4180 1.9716 33.5212 19.3637 31.1326 31.1162
2.0824 23.0 4370 1.9667 33.5156 19.3223 31.1023 31.0709
2.0824 24.0 4560 1.9735 33.6089 19.3842 31.1539 31.1419
2.0657 25.0 4750 1.9674 33.6317 19.4044 31.2361 31.2222
2.0657 26.0 4940 1.9617 33.745 19.5099 31.3061 31.2643
2.0657 27.0 5130 1.9613 33.7798 19.5496 31.3761 31.3356
2.0511 28.0 5320 1.9635 33.8568 19.594 31.4454 31.4141
2.0511 29.0 5510 1.9609 33.805 19.5962 31.393 31.3493
2.0377 30.0 5700 1.9610 33.7922 19.5758 31.3737 31.3428

Framework versions

  • Transformers 4.25.1
  • Pytorch 1.13.0
  • Datasets 2.8.0
  • Tokenizers 0.12.1