ModernBERT-base-2-contract-sections-classification-v4-50-1024

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: 0.4160
Accuracy Evaluate: 0.9353
Precision Evaluate: 0.9380
Recall Evaluate: 0.9337
F1 Evaluate: 0.9345
Accuracy Sklearn: 0.9353
Precision Sklearn: 0.9376
Recall Sklearn: 0.9353
F1 Sklearn: 0.9352
Acuracia Rotulo Objeto: 0.9897
Acuracia Rotulo Obrigacoes: 0.9613
Acuracia Rotulo Valor: 0.8653
Acuracia Rotulo Vigencia: 0.9738
Acuracia Rotulo Rescisao: 0.9280
Acuracia Rotulo Foro: 0.9962
Acuracia Rotulo Reajuste: 0.8897
Acuracia Rotulo Fiscalizacao: 0.8454
Acuracia Rotulo Publicacao: 0.9507
Acuracia Rotulo Pagamento: 0.8841
Acuracia Rotulo Casos Omissos: 0.9113
Acuracia Rotulo Sancoes: 0.9541
Acuracia Rotulo Dotacao Orcamentaria: 0.9890

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: 50

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
0.5171	1.0	1000	0.9144	0.732	0.7887	0.7494	0.7499	0.732	0.7735	0.732	0.7332	0.8822	0.6886	0.4556	0.6745	0.8033	0.8846	0.7509	0.7224	0.8276	0.4565	0.8227	0.9266	0.8462
0.2169	2.0	2000	0.6043	0.8325	0.8593	0.8461	0.8475	0.8325	0.8407	0.8325	0.8308	0.8926	0.8182	0.5587	0.7638	0.9003	0.9885	0.8434	0.7571	0.8818	0.8659	0.8571	0.8991	0.9725
0.1396	3.0	3000	0.4994	0.8602	0.8659	0.8764	0.8660	0.8602	0.8679	0.8602	0.8596	0.9050	0.8148	0.6447	0.8924	0.8061	0.9923	0.8932	0.8202	0.9655	0.8587	0.9015	0.9266	0.9725
0.1134	4.0	4000	0.4213	0.8925	0.8971	0.8970	0.8943	0.8925	0.8959	0.8925	0.8917	0.9483	0.8822	0.7249	0.9475	0.9363	0.9962	0.8505	0.7823	0.8966	0.8841	0.9015	0.9266	0.9835
0.09	5.0	5000	0.4038	0.9012	0.9057	0.9062	0.9046	0.9012	0.9031	0.9012	0.9010	0.9669	0.8687	0.7966	0.9318	0.9391	1.0	0.8648	0.8170	0.9163	0.8587	0.9064	0.9358	0.9780
0.0569	6.0	6000	0.4327	0.8925	0.8914	0.9055	0.8959	0.8925	0.8965	0.8925	0.8917	0.9773	0.7609	0.7966	0.9685	0.9307	0.9962	0.8470	0.8202	0.9507	0.8913	0.9015	0.9358	0.9945
0.0573	7.0	7000	0.3939	0.911	0.9142	0.9140	0.9122	0.911	0.9140	0.911	0.9108	0.9835	0.8939	0.7536	0.9633	0.9307	0.9962	0.8790	0.8675	0.9015	0.8696	0.9113	0.9541	0.9780
0.0433	8.0	8000	0.3762	0.92	0.9204	0.9193	0.9174	0.92	0.9238	0.92	0.9200	0.9814	0.9512	0.7822	0.9659	0.9197	0.9923	0.8790	0.8423	0.9015	0.8768	0.9261	0.9541	0.9780
0.0267	9.0	9000	0.3790	0.9215	0.9290	0.9192	0.9221	0.9215	0.9255	0.9215	0.9214	0.9855	0.9461	0.7851	0.9764	0.9335	0.9923	0.8754	0.8580	0.8867	0.8804	0.9064	0.9450	0.9780
0.0242	10.0	10000	0.4134	0.9153	0.9131	0.9174	0.9095	0.9153	0.9259	0.9153	0.9174	0.9876	0.9259	0.8252	0.9475	0.8476	0.9962	0.8577	0.8833	0.9113	0.8913	0.8966	0.9725	0.9835
0.0257	11.0	11000	0.3048	0.9365	0.9361	0.9346	0.9346	0.9365	0.9371	0.9365	0.9361	0.9835	0.9579	0.8596	0.9816	0.9529	0.9923	0.8826	0.8612	0.9163	0.8986	0.9163	0.9633	0.9835
0.0164	12.0	12000	0.3451	0.9313	0.9320	0.9292	0.9290	0.9313	0.9336	0.9313	0.9312	0.9773	0.9613	0.8768	0.9659	0.9446	0.9923	0.8719	0.8486	0.9163	0.8623	0.9212	0.9633	0.9780
0.0182	13.0	13000	0.3905	0.9275	0.9395	0.9254	0.9307	0.9275	0.9310	0.9275	0.9273	0.9917	0.9579	0.7908	0.9685	0.9363	0.9962	0.8897	0.8707	0.9409	0.8623	0.8966	0.9450	0.9835
0.0089	14.0	14000	0.4056	0.929	0.9340	0.9290	0.9301	0.929	0.9313	0.929	0.9286	0.9897	0.9579	0.8080	0.9685	0.9363	0.9962	0.8612	0.8454	0.9557	0.8913	0.9113	0.9725	0.9835
0.007	15.0	15000	0.3713	0.9313	0.9279	0.9321	0.9285	0.9313	0.9331	0.9313	0.9310	0.9773	0.9461	0.8539	0.9790	0.9280	0.9962	0.8790	0.8391	0.9310	0.9094	0.9163	0.9725	0.9890
0.0067	16.0	16000	0.3521	0.934	0.9322	0.9332	0.9316	0.934	0.9355	0.934	0.9340	0.9711	0.9596	0.8711	0.9764	0.9197	0.9885	0.9004	0.8644	0.9261	0.8913	0.9163	0.9633	0.9835
0.0053	17.0	17000	0.3731	0.9325	0.9337	0.9306	0.9305	0.9325	0.9346	0.9325	0.9322	0.9814	0.9630	0.8567	0.9816	0.9391	1.0	0.8861	0.8265	0.9163	0.8841	0.9163	0.9633	0.9835
0.005	18.0	18000	0.3707	0.9343	0.9337	0.9343	0.9329	0.9343	0.9359	0.9343	0.9341	0.9855	0.9512	0.8596	0.9711	0.9335	0.9962	0.8683	0.8675	0.9507	0.8877	0.9113	0.9633	1.0
0.0037	19.0	19000	0.3963	0.9343	0.9333	0.9337	0.9319	0.9343	0.9375	0.9343	0.9345	0.9897	0.9293	0.9054	0.9816	0.9307	0.9962	0.8683	0.8675	0.9458	0.8768	0.9015	0.9450	1.0
0.0026	20.0	20000	0.4162	0.9293	0.9343	0.9258	0.9283	0.9293	0.9324	0.9293	0.9291	0.9897	0.9613	0.8539	0.9790	0.9307	0.9923	0.8897	0.8328	0.9212	0.8551	0.9015	0.9450	0.9835
0.0021	21.0	21000	0.3894	0.9337	0.9364	0.9312	0.9326	0.9337	0.9357	0.9337	0.9336	0.9897	0.9630	0.8625	0.9659	0.9363	0.9962	0.8826	0.8517	0.9261	0.8877	0.9064	0.9541	0.9835
0.0018	22.0	22000	0.3775	0.9355	0.9354	0.9338	0.9334	0.9355	0.9372	0.9355	0.9353	0.9855	0.9596	0.8510	0.9764	0.9335	0.9962	0.8897	0.8770	0.9458	0.8804	0.9064	0.9541	0.9835
0.003	23.0	23000	0.3692	0.9353	0.9353	0.9337	0.9334	0.9353	0.9368	0.9353	0.9351	0.9855	0.9579	0.8711	0.9711	0.9335	0.9962	0.8861	0.8612	0.9261	0.8913	0.9113	0.9633	0.9835
0.0036	24.0	24000	0.4267	0.9285	0.9309	0.9271	0.9267	0.9285	0.9317	0.9285	0.9283	0.9793	0.9697	0.8539	0.9711	0.9224	0.9923	0.8861	0.8265	0.9163	0.8587	0.9113	0.9817	0.9835
0.0034	25.0	25000	0.4385	0.928	0.9338	0.9253	0.9273	0.928	0.9321	0.928	0.9281	0.9876	0.9613	0.8481	0.9711	0.9141	0.9962	0.8861	0.8423	0.9310	0.8659	0.8966	0.9450	0.9835
0.0021	26.0	26000	0.3957	0.9363	0.9364	0.9368	0.9349	0.9363	0.9388	0.9363	0.9364	0.9773	0.9428	0.8940	0.9790	0.9391	0.9962	0.8790	0.8675	0.9507	0.8841	0.8966	0.9725	1.0
0.0038	27.0	27000	0.4856	0.9227	0.9351	0.9218	0.9265	0.9227	0.9288	0.9227	0.9234	0.9917	0.9478	0.8109	0.9580	0.9169	0.9923	0.8612	0.8580	0.9360	0.8696	0.9015	0.9450	0.9945
0.0035	28.0	28000	0.4028	0.9353	0.9375	0.9325	0.9337	0.9353	0.9374	0.9353	0.9352	0.9917	0.9596	0.8596	0.9738	0.9280	0.9962	0.9146	0.8644	0.9212	0.8732	0.9113	0.9450	0.9835
0.0022	29.0	29000	0.4053	0.9367	0.9420	0.9345	0.9373	0.9367	0.9385	0.9367	0.9366	0.9876	0.9596	0.8539	0.9790	0.9529	0.9923	0.8861	0.8738	0.9458	0.8659	0.9064	0.9450	1.0
0.0017	30.0	30000	0.3755	0.936	0.9365	0.9347	0.9347	0.936	0.9375	0.936	0.9358	0.9855	0.9579	0.8510	0.9790	0.9307	0.9962	0.8932	0.8738	0.9507	0.8768	0.9113	0.9450	1.0
0.0025	31.0	31000	0.4102	0.9347	0.9403	0.9320	0.9347	0.9347	0.9375	0.9347	0.9348	0.9917	0.9630	0.8711	0.9685	0.9280	0.9962	0.8826	0.8580	0.9458	0.8768	0.9064	0.9450	0.9835
0.0011	32.0	32000	0.4056	0.935	0.9369	0.9330	0.9338	0.935	0.9371	0.935	0.9350	0.9876	0.9596	0.8653	0.9738	0.9280	0.9962	0.8861	0.8612	0.9507	0.8804	0.9064	0.9450	0.9890
0.0009	33.0	33000	0.4007	0.936	0.9380	0.9339	0.9348	0.936	0.9381	0.936	0.9360	0.9876	0.9596	0.8653	0.9738	0.9363	0.9962	0.8861	0.8644	0.9507	0.8804	0.9064	0.9450	0.9890
0.0021	34.0	34000	0.4073	0.9327	0.9368	0.9302	0.9322	0.9327	0.9351	0.9327	0.9326	0.9876	0.9596	0.8481	0.9816	0.9307	0.9962	0.8897	0.8549	0.9163	0.8768	0.9064	0.9450	1.0
0.0023	35.0	35000	0.3993	0.9355	0.9389	0.9345	0.9353	0.9355	0.9377	0.9355	0.9354	0.9835	0.9579	0.8797	0.9790	0.9252	0.9923	0.8861	0.8549	0.9507	0.8804	0.9064	0.9633	0.9890
0.0	36.0	36000	0.3990	0.9365	0.9383	0.9342	0.9353	0.9365	0.9381	0.9365	0.9364	0.9897	0.9596	0.8625	0.9790	0.9363	0.9962	0.8897	0.8644	0.9507	0.8768	0.9064	0.9450	0.9890
0.001	37.0	37000	0.3950	0.9393	0.9406	0.9376	0.9383	0.9393	0.9405	0.9393	0.9391	0.9897	0.9596	0.8625	0.9790	0.9363	0.9923	0.9039	0.8801	0.9507	0.8804	0.9113	0.9541	0.9890
0.001	38.0	38000	0.3969	0.9373	0.9400	0.9349	0.9366	0.9373	0.9388	0.9373	0.9372	0.9876	0.9613	0.8596	0.9790	0.9363	0.9962	0.8932	0.8707	0.9458	0.8841	0.9064	0.9450	0.9890
0.0	39.0	39000	0.4129	0.9367	0.9370	0.9356	0.9352	0.9367	0.9385	0.9367	0.9366	0.9876	0.9596	0.8596	0.9790	0.9280	0.9923	0.8897	0.8770	0.9507	0.8804	0.9064	0.9633	0.9890
0.002	40.0	40000	0.4130	0.9335	0.9346	0.9332	0.9325	0.9335	0.9354	0.9335	0.9332	0.9835	0.9613	0.8596	0.9738	0.9280	0.9923	0.8897	0.8328	0.9507	0.8877	0.9113	0.9725	0.9890
0.0007	41.0	41000	0.3864	0.939	0.9441	0.9365	0.9395	0.939	0.9402	0.939	0.9388	0.9876	0.9630	0.8625	0.9764	0.9557	0.9962	0.9039	0.8612	0.9458	0.8768	0.9113	0.9450	0.9890
0.0006	42.0	42000	0.4130	0.936	0.9388	0.9349	0.9359	0.936	0.9375	0.936	0.9358	0.9876	0.9596	0.8367	0.9790	0.9418	0.9962	0.8897	0.8738	0.9507	0.8804	0.9064	0.9633	0.9890
0.0007	43.0	43000	0.4165	0.9367	0.9388	0.9346	0.9358	0.9367	0.9384	0.9367	0.9366	0.9897	0.9596	0.8510	0.9790	0.9363	0.9962	0.8897	0.8770	0.9507	0.8804	0.9064	0.9450	0.9890
0.001	44.0	44000	0.4147	0.9357	0.9370	0.9349	0.9346	0.9357	0.9378	0.9357	0.9356	0.9897	0.9613	0.8596	0.9738	0.9307	0.9923	0.8897	0.8612	0.9507	0.8768	0.9064	0.9725	0.9890
0.0017	45.0	45000	0.4130	0.9363	0.9371	0.9342	0.9346	0.9363	0.9382	0.9363	0.9362	0.9897	0.9596	0.8653	0.9764	0.9307	0.9962	0.8897	0.8612	0.9507	0.8804	0.9113	0.9450	0.9890
0.0	46.0	46000	0.4122	0.936	0.9409	0.9336	0.9361	0.936	0.9381	0.936	0.9359	0.9897	0.9613	0.8625	0.9764	0.9307	0.9962	0.8897	0.8644	0.9507	0.8804	0.9064	0.9450	0.9835
0.001	47.0	47000	0.4153	0.9353	0.9377	0.9340	0.9344	0.9353	0.9376	0.9353	0.9352	0.9897	0.9613	0.8653	0.9738	0.9280	0.9962	0.8897	0.8454	0.9507	0.8841	0.9113	0.9633	0.9835
0.0015	48.0	48000	0.4161	0.9355	0.9382	0.9340	0.9347	0.9355	0.9379	0.9355	0.9354	0.9897	0.9613	0.8682	0.9738	0.9280	0.9962	0.8897	0.8454	0.9507	0.8841	0.9113	0.9541	0.9890
0.0008	49.0	49000	0.4165	0.935	0.9378	0.9335	0.9343	0.935	0.9373	0.935	0.9349	0.9897	0.9613	0.8653	0.9738	0.9280	0.9962	0.8897	0.8454	0.9507	0.8804	0.9113	0.9541	0.9890
0.0005	50.0	50000	0.4160	0.9353	0.9380	0.9337	0.9345	0.9353	0.9376	0.9353	0.9352	0.9897	0.9613	0.8653	0.9738	0.9280	0.9962	0.8897	0.8454	0.9507	0.8841	0.9113	0.9541	0.9890

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-v4-50-1024

ModernBERT-base-2-contract-sections-classification-v4-50-1024

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Seed this model