ModernBERT-base-2-contract-sections-classification-v4-10-1024

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

Loss: 0.4077
Accuracy Evaluate: 0.9
Precision Evaluate: 0.9119
Recall Evaluate: 0.9038
F1 Evaluate: 0.9061
Accuracy Sklearn: 0.9
Precision Sklearn: 0.9031
Recall Sklearn: 0.9
F1 Sklearn: 0.8996
Acuracia Rotulo Objeto: 0.9731
Acuracia Rotulo Obrigacoes: 0.8771
Acuracia Rotulo Valor: 0.7192
Acuracia Rotulo Vigencia: 0.9528
Acuracia Rotulo Rescisao: 0.9446
Acuracia Rotulo Foro: 0.9923
Acuracia Rotulo Reajuste: 0.8683
Acuracia Rotulo Fiscalizacao: 0.8423
Acuracia Rotulo Publicacao: 0.9064
Acuracia Rotulo Pagamento: 0.8623
Acuracia Rotulo Casos Omissos: 0.8966
Acuracia Rotulo Sancoes: 0.9358
Acuracia Rotulo Dotacao Orcamentaria: 0.9780

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
0.529	1.0	1000	0.9404	0.7238	0.7893	0.7401	0.7414	0.7238	0.7739	0.7238	0.7242	0.8946	0.6751	0.4384	0.6325	0.8283	0.8769	0.7331	0.7413	0.8621	0.4094	0.8030	0.9083	0.8187
0.2306	2.0	2000	0.6074	0.841	0.8638	0.8549	0.8543	0.841	0.8481	0.841	0.8388	0.8781	0.8249	0.5415	0.8556	0.9030	0.9846	0.8505	0.7445	0.9409	0.8514	0.8670	0.8991	0.9725
0.1523	3.0	3000	0.5212	0.8562	0.8655	0.8700	0.8631	0.8562	0.8616	0.8562	0.8546	0.8946	0.8418	0.5903	0.8871	0.8393	0.9962	0.8932	0.7823	0.9606	0.8478	0.8966	0.9083	0.9725
0.1231	4.0	4000	0.4573	0.8788	0.8925	0.8873	0.8872	0.8788	0.8828	0.8788	0.8779	0.9545	0.8367	0.6619	0.9370	0.9030	0.9962	0.8612	0.7950	0.9458	0.8659	0.8916	0.9083	0.9780
0.101	5.0	5000	0.4354	0.8872	0.8960	0.8983	0.8954	0.8872	0.8905	0.8872	0.8871	0.9628	0.8182	0.7593	0.8793	0.9391	0.9962	0.8719	0.8170	0.9557	0.8587	0.9064	0.9358	0.9780
0.0687	6.0	6000	0.4196	0.8962	0.9050	0.9042	0.9030	0.8962	0.8985	0.8962	0.8957	0.9587	0.8468	0.7249	0.9554	0.9446	0.9885	0.8790	0.8265	0.9557	0.8587	0.9015	0.9358	0.9780
0.0747	7.0	7000	0.4255	0.8928	0.9067	0.9009	0.9017	0.8928	0.8966	0.8928	0.8924	0.9649	0.8316	0.7221	0.9554	0.9446	0.9923	0.8648	0.8360	0.9360	0.8478	0.8966	0.9358	0.9835
0.0627	8.0	8000	0.4168	0.8965	0.9073	0.9028	0.9031	0.8965	0.8999	0.8965	0.8961	0.9752	0.8384	0.7393	0.9580	0.9474	0.9923	0.8577	0.8391	0.9163	0.8623	0.8966	0.9358	0.9780
0.0485	9.0	9000	0.4104	0.898	0.9089	0.9041	0.9048	0.898	0.9018	0.898	0.8979	0.9731	0.8418	0.7536	0.9501	0.9446	0.9923	0.8683	0.8486	0.9064	0.8587	0.9015	0.9358	0.9780
0.048	10.0	10000	0.4077	0.9	0.9119	0.9038	0.9061	0.9	0.9031	0.9	0.8996	0.9731	0.8771	0.7192	0.9528	0.9446	0.9923	0.8683	0.8423	0.9064	0.8623	0.8966	0.9358	0.9780

Framework versions

Transformers 4.49.0
Pytorch 2.6.0+cu124
Datasets 3.3.0
Tokenizers 0.21.0

marcelovidigal/ModernBERT-base-2-contract-sections-classification-v4-10-1024

ModernBERT-base-2-contract-sections-classification-v4-10-1024

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种