Edit model card

balanced-augmented-roberta-large-gest-pred-seqeval-partialmatch-2

This model is a fine-tuned version of roberta-large on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3718
  • Precision: 0.9171
  • Recall: 0.9183
  • F1: 0.9144
  • Accuracy: 0.9173

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: 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: 20

Training results

Training Loss Epoch Step Validation Loss Precision Recall F1 Accuracy
2.8924 1.0 52 2.1510 0.3032 0.2572 0.2464 0.4120
1.8492 2.0 104 1.3548 0.5880 0.5350 0.5046 0.6440
1.1692 3.0 156 0.8621 0.7443 0.6770 0.6775 0.7468
0.7452 4.0 208 0.6391 0.7779 0.7919 0.7691 0.8159
0.5185 5.0 260 0.6401 0.8197 0.8000 0.7868 0.8109
0.3536 6.0 312 0.4251 0.8808 0.8684 0.8675 0.8820
0.2411 7.0 364 0.4748 0.8709 0.8659 0.8613 0.8800
0.1762 8.0 416 0.3809 0.8991 0.8721 0.8812 0.8971
0.1377 9.0 468 0.3977 0.9062 0.8950 0.8953 0.9022
0.1026 10.0 520 0.4637 0.9068 0.8887 0.8897 0.8951
0.0763 11.0 572 0.4210 0.9079 0.9066 0.9020 0.9012
0.0554 12.0 624 0.4950 0.8962 0.8837 0.8809 0.8850
0.0419 13.0 676 0.4643 0.9043 0.9007 0.8969 0.8961
0.0358 14.0 728 0.3718 0.9171 0.9183 0.9144 0.9173
0.0264 15.0 780 0.4456 0.9349 0.9042 0.9132 0.9123
0.0178 16.0 832 0.4296 0.9362 0.9169 0.9227 0.9193
0.015 17.0 884 0.3900 0.9302 0.9235 0.9240 0.9254
0.0105 18.0 936 0.4335 0.9284 0.9161 0.9181 0.9168
0.0116 19.0 988 0.4426 0.9285 0.9138 0.9166 0.9138
0.0104 20.0 1040 0.4410 0.9276 0.9141 0.9163 0.9138

Framework versions

  • Transformers 4.28.1
  • Pytorch 2.0.0+cu118
  • Datasets 2.11.0
  • Tokenizers 0.13.3

LICENSE

Copyright (c) 2014, Universidad Carlos III de Madrid. Todos los derechos reservados. Este software es propiedad de la Universidad Carlos III de Madrid, grupo de investigación Robots Sociales. La Universidad Carlos III de Madrid es titular en exclusiva de los derechos de propiedad intelectual de este software. Queda prohibido cualquier uso indebido o no autorizado, entre estos, a título enunciativo pero no limitativo, la reproducción, fijación, distribución, comunicación pública, ingeniería inversa y/o transformación sobre dicho software, ya sea total o parcialmente, siendo el responsable del uso indebido o no autorizado también responsable de las consecuencias legales que pudieran derivarse de sus actos.

Downloads last month
8
Inference Examples
Inference API (serverless) is not available, repository is disabled.

Dataset used to train Jsevisal/balanced-augmented-roberta-large-gest-pred-seqeval-partialmatch-2