ModernBERT-base-2-contract-sections-classification-v4-10-512

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: 0.3151
Accuracy Evaluate: 0.9207
Precision Evaluate: 0.9230
Recall Evaluate: 0.9313
F1 Evaluate: 0.9262
Accuracy Sklearn: 0.9207
Precision Sklearn: 0.9228
Recall Sklearn: 0.9207
F1 Sklearn: 0.9206
Acuracia Rotulo Objeto: 0.9669
Acuracia Rotulo Obrigacoes: 0.8199
Acuracia Rotulo Valor: 0.8567
Acuracia Rotulo Vigencia: 0.9580
Acuracia Rotulo Rescisao: 0.9446
Acuracia Rotulo Foro: 0.9885
Acuracia Rotulo Reajuste: 0.9680
Acuracia Rotulo Fiscalizacao: 0.8486
Acuracia Rotulo Publicacao: 0.9901
Acuracia Rotulo Pagamento: 0.9348
Acuracia Rotulo Casos Omissos: 0.9113
Acuracia Rotulo Sancoes: 0.9358
Acuracia Rotulo Dotacao Orcamentaria: 0.9835

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
0.4789	1.0	1000	0.8779	0.7163	0.7889	0.7364	0.7408	0.7163	0.7689	0.7163	0.7179	0.9050	0.6313	0.4269	0.7690	0.8726	0.8269	0.3345	0.5741	0.8670	0.7065	0.8276	0.9358	0.8956
0.2075	2.0	2000	0.5023	0.8558	0.8682	0.8744	0.8685	0.8558	0.8605	0.8558	0.8546	0.8822	0.7424	0.6361	0.9423	0.9086	0.9885	0.8683	0.7666	0.9606	0.9203	0.8818	0.9083	0.9615
0.1423	3.0	3000	0.4114	0.884	0.8792	0.8990	0.8856	0.884	0.8884	0.884	0.8840	0.8822	0.8081	0.7507	0.9475	0.8837	0.9885	0.9502	0.7981	0.9754	0.9275	0.8867	0.9266	0.9615
0.1048	4.0	4000	0.3871	0.8925	0.8996	0.9083	0.9020	0.8925	0.8961	0.8925	0.8919	0.9008	0.7677	0.7622	0.9554	0.9695	0.9962	0.9181	0.8265	0.9754	0.9493	0.8916	0.9174	0.9780
0.0871	5.0	5000	0.3639	0.9028	0.9077	0.9161	0.9104	0.9028	0.9067	0.9028	0.9028	0.9587	0.7694	0.8539	0.9396	0.9335	0.9923	0.9609	0.8265	0.9901	0.8877	0.9015	0.9174	0.9780
0.0604	6.0	6000	0.3634	0.908	0.9105	0.9223	0.9148	0.908	0.9123	0.908	0.9080	0.9339	0.7626	0.8682	0.9711	0.9280	0.9923	0.9644	0.8549	0.9852	0.9384	0.9015	0.9174	0.9725
0.0611	7.0	7000	0.3359	0.9125	0.9202	0.9259	0.9216	0.9125	0.9160	0.9125	0.9123	0.9483	0.7845	0.8166	0.9685	0.9612	0.9885	0.9644	0.8580	0.9901	0.9457	0.8966	0.9358	0.9780
0.0514	8.0	8000	0.3315	0.92	0.9235	0.9299	0.9258	0.92	0.9222	0.92	0.9198	0.9628	0.8215	0.8395	0.9659	0.9612	0.9923	0.9466	0.8549	0.9901	0.9384	0.9015	0.9358	0.9780
0.0398	9.0	9000	0.3208	0.9187	0.9204	0.9290	0.9235	0.9187	0.9214	0.9187	0.9187	0.9711	0.8199	0.8567	0.9528	0.9335	0.9923	0.9680	0.8486	0.9901	0.9239	0.9064	0.9358	0.9780
0.0396	10.0	10000	0.3151	0.9207	0.9230	0.9313	0.9262	0.9207	0.9228	0.9207	0.9206	0.9669	0.8199	0.8567	0.9580	0.9446	0.9885	0.9680	0.8486	0.9901	0.9348	0.9113	0.9358	0.9835

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-v4-10-512

ModernBERT-base-2-contract-sections-classification-v4-10-512

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model