bert-base-multilingual-cased-2-contract-sections-classification-v4-10

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

Loss: 0.1797
Accuracy Evaluate: 0.9603
Precision Evaluate: 0.9530
Recall Evaluate: 0.9655
F1 Evaluate: 0.9578
Accuracy Sklearn: 0.9603
Precision Sklearn: 0.9623
Recall Sklearn: 0.9603
F1 Sklearn: 0.9605
Acuracia Rotulo Objeto: 0.9917
Acuracia Rotulo Obrigacoes: 0.9259
Acuracia Rotulo Valor: 0.9284
Acuracia Rotulo Vigencia: 0.9843
Acuracia Rotulo Rescisao: 0.9169
Acuracia Rotulo Foro: 1.0
Acuracia Rotulo Reajuste: 0.9964
Acuracia Rotulo Fiscalizacao: 0.9180
Acuracia Rotulo Publicacao: 1.0
Acuracia Rotulo Pagamento: 0.9783
Acuracia Rotulo Casos Omissos: 0.9212
Acuracia Rotulo Sancoes: 0.9908
Acuracia Rotulo Dotacao Orcamentaria: 1.0

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
1.5525	1.0	1000	1.2169	0.7772	0.7719	0.7417	0.7406	0.7772	0.7416	0.7772	0.7485	0.9483	0.9209	0.8940	0.6719	0.7507	0.9192	0.8327	0.7886	0.9507	0.0	0.9113	0.3945	0.6593
0.4981	2.0	2000	0.4333	0.9337	0.9326	0.9382	0.9347	0.9337	0.9358	0.9337	0.9340	0.9773	0.8586	0.9255	0.9764	0.9418	0.9308	0.9680	0.8896	0.9951	0.9130	0.9212	0.8991	1.0
0.2197	3.0	3000	0.2467	0.9463	0.9473	0.9527	0.9494	0.9463	0.9474	0.9463	0.9463	0.9793	0.8872	0.9284	0.9790	0.9418	0.9385	0.9964	0.8896	0.9951	0.9565	0.9212	0.9725	1.0
0.1351	4.0	4000	0.2067	0.9503	0.9516	0.9565	0.9535	0.9503	0.9515	0.9503	0.9503	0.9835	0.8889	0.9284	0.9843	0.9418	0.9577	0.9751	0.9148	0.9951	0.9710	0.9212	0.9725	1.0
0.0967	5.0	5000	0.2049	0.951	0.9523	0.9589	0.9550	0.951	0.9525	0.951	0.9511	0.9814	0.8788	0.9341	0.9790	0.9418	0.9885	0.9751	0.9180	0.9951	0.9710	0.9212	0.9817	1.0
0.0697	6.0	6000	0.1977	0.953	0.9535	0.9612	0.9567	0.953	0.9544	0.953	0.9531	0.9814	0.8822	0.9284	0.9869	0.9418	0.9885	0.9786	0.9180	1.0	0.9783	0.9212	0.9908	1.0
0.0711	7.0	7000	0.1787	0.958	0.9516	0.9630	0.9561	0.958	0.9596	0.958	0.9582	0.9855	0.9276	0.9284	0.9869	0.9252	1.0	0.9751	0.9085	0.9951	0.9746	0.9212	0.9908	1.0
0.0614	8.0	8000	0.1885	0.9573	0.9511	0.9634	0.9561	0.9573	0.9591	0.9573	0.9575	0.9917	0.9091	0.9284	0.9843	0.9280	1.0	0.9751	0.9180	1.0	0.9783	0.9212	0.9908	1.0
0.0576	9.0	9000	0.1802	0.959	0.9510	0.9638	0.9559	0.959	0.9611	0.959	0.9593	0.9917	0.9310	0.9284	0.9843	0.9114	1.0	0.9751	0.9180	1.0	0.9783	0.9212	0.9908	1.0
0.0512	10.0	10000	0.1797	0.9603	0.9530	0.9655	0.9578	0.9603	0.9623	0.9603	0.9605	0.9917	0.9259	0.9284	0.9843	0.9169	1.0	0.9964	0.9180	1.0	0.9783	0.9212	0.9908	1.0

Framework versions

Transformers 4.51.3
Pytorch 2.7.0+cu126
Datasets 3.5.1
Tokenizers 0.21.1

marcelovidigal/bert-base-multilingual-cased-2-contract-sections-classification-v4-10

bert-base-multilingual-cased-2-contract-sections-classification-v4-10

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model