balanced-augmented-distilbert-base-gest-pred-seqeval-partialmatch-2

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

  • Loss: 0.3958
  • Precision: 0.9273
  • Recall: 0.9067
  • F1: 0.9116
  • Accuracy: 0.9005

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
3.1391 1.0 52 2.5842 0.1830 0.1356 0.1249 0.3087
2.1708 2.0 104 1.7976 0.3943 0.3943 0.3622 0.5473
1.5993 3.0 156 1.3522 0.5329 0.4921 0.4697 0.6482
1.2237 4.0 208 1.1070 0.6701 0.5833 0.5820 0.6923
0.9516 5.0 260 0.9920 0.7506 0.6840 0.6820 0.7501
0.7364 6.0 312 0.7888 0.8171 0.7327 0.7350 0.7815
0.5816 7.0 364 0.6753 0.8510 0.7845 0.7961 0.8256
0.4507 8.0 416 0.5905 0.8807 0.8242 0.8400 0.8550
0.3589 9.0 468 0.5305 0.9021 0.8667 0.8746 0.8702
0.2875 10.0 520 0.5081 0.9176 0.8788 0.8911 0.8834
0.2268 11.0 572 0.4599 0.9133 0.8879 0.8939 0.8863
0.1875 12.0 624 0.4347 0.9224 0.8946 0.9025 0.8942
0.1608 13.0 676 0.4497 0.9186 0.8846 0.8956 0.8873
0.1356 14.0 728 0.4271 0.9242 0.8951 0.9038 0.8932
0.1197 15.0 780 0.3958 0.9273 0.9067 0.9116 0.9005
0.1015 16.0 832 0.4060 0.9285 0.9013 0.9095 0.8991
0.0896 17.0 884 0.4023 0.9300 0.9114 0.9157 0.9040
0.0847 18.0 936 0.4041 0.9296 0.9077 0.9133 0.9005
0.0811 19.0 988 0.4069 0.9291 0.9089 0.9133 0.9005
0.0752 20.0 1040 0.4086 0.9286 0.9075 0.9129 0.8996

Framework versions

  • Transformers 4.27.3
  • Pytorch 1.13.1+cu116
  • Datasets 2.10.1
  • Tokenizers 0.13.2

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
17
Safetensors
Model size
65.2M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.

Dataset used to train Jsevisal/balanced-augmented-distilbert-base-gest-pred-seqeval-partialmatch-2