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

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

• Loss: 0.3427
• Precision: 0.9361
• Recall: 0.9389
• F1: 0.9320
• Accuracy: 0.9260

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
2.9298	1.0	52	2.3822	0.2363	0.1557	0.1575	0.3204
1.9949	2.0	104	1.5817	0.5566	0.5259	0.4978	0.5958
1.3242	3.0	156	1.0665	0.6572	0.6680	0.6417	0.7124
0.8143	4.0	208	0.7375	0.8047	0.8024	0.7876	0.7972
0.4744	5.0	260	0.5433	0.8598	0.8570	0.8434	0.8476
0.2876	6.0	312	0.4301	0.8945	0.9034	0.8911	0.8868
0.1784	7.0	364	0.5261	0.9056	0.8915	0.8866	0.8711
0.1103	8.0	416	0.4828	0.9169	0.9172	0.9066	0.8917
0.076	9.0	468	0.3915	0.9116	0.9075	0.9016	0.8956
0.053	10.0	520	0.3593	0.9167	0.9299	0.9177	0.9143
0.0364	11.0	572	0.3427	0.9361	0.9389	0.9320	0.9260
0.028	12.0	624	0.3638	0.9275	0.9327	0.9253	0.9162
0.0195	13.0	676	0.3486	0.9268	0.9416	0.9298	0.9216
0.0156	14.0	728	0.4049	0.9204	0.9256	0.9156	0.9030
0.0146	15.0	780	0.3894	0.9267	0.9311	0.9224	0.9152
0.01	16.0	832	0.3661	0.9268	0.9342	0.9248	0.9201
0.0082	17.0	884	0.3897	0.9243	0.9293	0.9197	0.9133
0.0076	18.0	936	0.3723	0.9254	0.9353	0.9250	0.9192
0.0069	19.0	988	0.3841	0.9277	0.9322	0.9236	0.9157
0.0075	20.0	1040	0.3825	0.9273	0.9325	0.9236	0.9157

Framework versions

• Transformers 4.27.4
• Pytorch 1.13.1+cu116
• Datasets 2.11.0
• 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.