albert-base-v2-2-contract-sections-classification-v4-10

This model is a fine-tuned version of albert/albert-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 0.8591
Accuracy Evaluate: 0.7843
Precision Evaluate: 0.8045
Recall Evaluate: 0.7910
F1 Evaluate: 0.7930
Accuracy Sklearn: 0.7843
Precision Sklearn: 0.7973
Recall Sklearn: 0.7843
F1 Sklearn: 0.7854
Acuracia Rotulo Objeto: 0.8698
Acuracia Rotulo Obrigacoes: 0.8670
Acuracia Rotulo Valor: 0.6046
Acuracia Rotulo Vigencia: 0.5984
Acuracia Rotulo Rescisao: 0.7839
Acuracia Rotulo Foro: 0.9
Acuracia Rotulo Reajuste: 0.8185
Acuracia Rotulo Fiscalizacao: 0.6656
Acuracia Rotulo Publicacao: 0.8227
Acuracia Rotulo Pagamento: 0.7717
Acuracia Rotulo Casos Omissos: 0.8522
Acuracia Rotulo Sancoes: 0.8716
Acuracia Rotulo Dotacao Orcamentaria: 0.8571

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: 1e-06
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy Evaluate	Precision Evaluate	Recall Evaluate	F1 Evaluate	Accuracy Sklearn	Precision Sklearn	Recall Sklearn	F1 Sklearn	Acuracia Rotulo Objeto	Acuracia Rotulo Obrigacoes	Acuracia Rotulo Valor	Acuracia Rotulo Vigencia	Acuracia Rotulo Rescisao	Acuracia Rotulo Foro	Acuracia Rotulo Reajuste	Acuracia Rotulo Fiscalizacao	Acuracia Rotulo Publicacao	Acuracia Rotulo Pagamento	Acuracia Rotulo Casos Omissos	Acuracia Rotulo Sancoes	Acuracia Rotulo Dotacao Orcamentaria
1.7695	1.0	1000	1.9528	0.4447	0.6693	0.3931	0.3947	0.4447	0.6282	0.4447	0.4028	0.9029	0.8148	0.4069	0.0577	0.2521	0.8692	0.2064	0.0789	0.5025	0.2283	0.4335	0.3303	0.0275
1.2261	2.0	2000	1.6110	0.5755	0.6694	0.5569	0.5614	0.5755	0.6584	0.5755	0.5645	0.8244	0.8013	0.4097	0.3963	0.3601	0.9077	0.6299	0.1956	0.6059	0.6051	0.5714	0.6514	0.2802
0.9134	3.0	3000	1.3841	0.6538	0.7003	0.6524	0.6478	0.6538	0.6991	0.6538	0.6473	0.8161	0.8182	0.4585	0.4724	0.4377	0.9038	0.7794	0.3218	0.7438	0.7428	0.7635	0.7339	0.4890
0.7022	4.0	4000	1.2144	0.6935	0.7370	0.6945	0.6965	0.6935	0.7265	0.6935	0.6904	0.8244	0.8333	0.4470	0.5407	0.6842	0.8923	0.8043	0.3596	0.7488	0.7210	0.6453	0.8349	0.6923
0.5597	5.0	5000	1.0738	0.7288	0.7560	0.7406	0.7386	0.7288	0.7494	0.7288	0.7286	0.8574	0.7845	0.5358	0.5171	0.7147	0.9	0.8292	0.5110	0.7783	0.7355	0.8128	0.8716	0.7802
0.4429	6.0	6000	0.9868	0.7552	0.7772	0.7641	0.7626	0.7552	0.7731	0.7552	0.7556	0.8678	0.8350	0.5874	0.5328	0.7368	0.8846	0.8399	0.5584	0.8227	0.7790	0.8276	0.8807	0.7802
0.389	7.0	7000	0.9236	0.7615	0.7823	0.7701	0.7683	0.7615	0.7763	0.7615	0.7603	0.8616	0.8620	0.6017	0.5538	0.7479	0.8962	0.8185	0.5142	0.8227	0.7681	0.8473	0.8716	0.8462
0.3341	8.0	8000	0.8949	0.776	0.7961	0.7833	0.7845	0.776	0.7900	0.776	0.7771	0.8781	0.8519	0.6017	0.5722	0.7729	0.8962	0.8078	0.6656	0.8227	0.7536	0.8424	0.8716	0.8462
0.3099	9.0	9000	0.8650	0.7805	0.8039	0.7876	0.7905	0.7805	0.7956	0.7805	0.7819	0.8822	0.8552	0.6160	0.5748	0.7756	0.8962	0.8114	0.6593	0.8227	0.7754	0.8522	0.8716	0.8462
0.3016	10.0	10000	0.8591	0.7843	0.8045	0.7910	0.7930	0.7843	0.7973	0.7843	0.7854	0.8698	0.8670	0.6046	0.5984	0.7839	0.9	0.8185	0.6656	0.8227	0.7717	0.8522	0.8716	0.8571

Framework versions

Transformers 4.48.3
Pytorch 2.6.0+cu124
Datasets 3.3.0
Tokenizers 0.21.0

marcelovidigal/albert-base-v2-2-contract-sections-classification-v4-10

albert-base-v2-2-contract-sections-classification-v4-10

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种