balanced-augmented-bert-gest-pred-seqeval-partialmatch-2

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

• Loss: 0.3649
• Precision: 0.9322
• Recall: 0.9236
• F1: 0.9235
• Accuracy: 0.9138

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
3.1239	1.0	52	2.5758	0.1906	0.1293	0.1175	0.3121
2.1363	2.0	104	1.7477	0.4478	0.4505	0.4151	0.5576
1.5182	3.0	156	1.3113	0.7051	0.6014	0.5633	0.6507
1.0948	4.0	208	0.9763	0.7361	0.6854	0.6695	0.7183
0.7591	5.0	260	0.7834	0.7900	0.7905	0.7726	0.7869
0.5168	6.0	312	0.5764	0.8775	0.8569	0.8479	0.8550
0.3609	7.0	364	0.5130	0.9055	0.8857	0.8815	0.8760
0.2538	8.0	416	0.4872	0.9106	0.8828	0.8865	0.8805
0.1898	9.0	468	0.3937	0.9219	0.9070	0.9076	0.8996
0.1343	10.0	520	0.3897	0.9271	0.9016	0.9095	0.9010
0.1053	11.0	572	0.3900	0.9309	0.9085	0.9143	0.9030
0.0788	12.0	624	0.3649	0.9322	0.9236	0.9235	0.9138
0.0643	13.0	676	0.4147	0.9293	0.9073	0.9122	0.9045
0.0501	14.0	728	0.4788	0.9369	0.9205	0.9200	0.9054
0.0424	15.0	780	0.4003	0.9346	0.9180	0.9197	0.9094
0.0376	16.0	832	0.3686	0.9373	0.9261	0.9274	0.9182
0.0317	17.0	884	0.4025	0.9360	0.9199	0.9223	0.9098
0.0288	18.0	936	0.4484	0.9406	0.9212	0.9239	0.9098
0.0282	19.0	988	0.4004	0.9377	0.9207	0.9238	0.9094
0.0256	20.0	1040	0.4046	0.9383	0.9205	0.9237	0.9098

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.