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: 0.2447
Accuracy Evaluate: 0.9625
Precision Evaluate: 0.9524
Recall Evaluate: 0.9635
F1 Evaluate: 0.9558
Accuracy Sklearn: 0.9625
Precision Sklearn: 0.9657
Recall Sklearn: 0.9625
F1 Sklearn: 0.9631
Acuracia Rotulo Objeto: 0.9917
Acuracia Rotulo Obrigacoes: 0.9663
Acuracia Rotulo Valor: 0.9427
Acuracia Rotulo Vigencia: 0.9869
Acuracia Rotulo Rescisao: 0.9114
Acuracia Rotulo Foro: 0.9423
Acuracia Rotulo Reajuste: 0.9715
Acuracia Rotulo Fiscalizacao: 0.9306
Acuracia Rotulo Publicacao: 1.0
Acuracia Rotulo Pagamento: 0.9746
Acuracia Rotulo Casos Omissos: 0.9261
Acuracia Rotulo Sancoes: 0.9817
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: 2
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 32
optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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
2.2161	1.0	500	1.8294	0.568	0.6269	0.5325	0.4985	0.568	0.6153	0.568	0.5258	0.8740	0.8822	0.6361	0.5932	0.3823	0.4385	0.3950	0.2397	0.9163	0.0	0.8621	0.6972	0.0055
1.2279	2.0	1000	0.8815	0.8842	0.9052	0.8856	0.8882	0.8842	0.8958	0.8842	0.8825	0.9607	0.9394	0.9284	0.7375	0.9280	0.9385	0.9573	0.7319	0.9803	0.6087	0.9310	0.9266	0.9451
0.5798	3.0	1500	0.4362	0.9345	0.9364	0.9424	0.9383	0.9345	0.9376	0.9345	0.9346	0.9649	0.8485	0.9284	0.9764	0.9529	0.9385	0.9893	0.9022	0.9852	0.8841	0.9360	0.9450	1.0
0.3013	4.0	2000	0.2795	0.9415	0.9357	0.9502	0.9407	0.9415	0.9458	0.9415	0.9420	0.9731	0.8603	0.9312	0.9948	0.9335	0.9385	0.9929	0.9148	0.9852	0.9058	0.9409	0.9817	1.0
0.1844	5.0	2500	0.2176	0.949	0.9413	0.9549	0.9462	0.949	0.9521	0.949	0.9495	0.9835	0.9040	0.9484	0.9895	0.9197	0.9385	0.9680	0.9148	0.9951	0.9203	0.9409	0.9908	1.0
0.1245	6.0	3000	0.1780	0.9605	0.9536	0.9630	0.9572	0.9605	0.9619	0.9605	0.9606	0.9814	0.9545	0.9312	0.9869	0.9335	0.9385	0.9964	0.9148	0.9951	0.9638	0.9409	0.9817	1.0
0.0911	7.0	3500	0.1572	0.9643	0.9550	0.9658	0.9582	0.9643	0.9675	0.9643	0.9649	0.9938	0.9697	0.9427	0.9816	0.9169	0.9385	0.9964	0.9180	0.9951	0.9710	0.9409	0.9908	1.0
0.0708	8.0	4000	0.1551	0.963	0.9530	0.9661	0.9572	0.963	0.9664	0.963	0.9637	0.9855	0.9596	0.9398	0.9869	0.9114	0.9346	0.9964	0.9338	0.9951	0.9710	0.9458	1.0	1.0
0.0591	9.0	4500	0.1480	0.9617	0.9528	0.9620	0.9556	0.9617	0.9644	0.9617	0.9622	0.9917	0.9697	0.9484	0.9869	0.9114	0.9423	0.9680	0.9180	0.9951	0.9710	0.9310	0.9725	1.0
0.0489	10.0	5000	0.1541	0.962	0.9528	0.9627	0.9556	0.962	0.9652	0.962	0.9626	0.9917	0.9697	0.9484	0.9869	0.9114	0.9423	0.9680	0.9180	0.9951	0.9710	0.9310	0.9817	1.0
0.0427	11.0	5500	0.1543	0.9635	0.9538	0.9656	0.9576	0.9635	0.9665	0.9635	0.9641	0.9917	0.9714	0.9456	0.9843	0.9114	0.9423	0.9644	0.9180	0.9951	0.9783	0.9507	1.0	1.0
0.0351	12.0	6000	0.1583	0.964	0.9552	0.9660	0.9587	0.964	0.9666	0.964	0.9645	0.9917	0.9714	0.9513	0.9869	0.9114	0.9385	0.9644	0.9180	1.0	0.9783	0.9458	1.0	1.0
0.03	13.0	6500	0.1687	0.961	0.9512	0.9623	0.9546	0.961	0.9642	0.961	0.9617	0.9917	0.9663	0.9513	0.9843	0.9114	0.9423	0.9466	0.9180	1.0	0.9710	0.9458	0.9817	1.0
0.029	14.0	7000	0.1652	0.9627	0.9544	0.9639	0.9578	0.9627	0.9647	0.9627	0.9631	0.9938	0.9697	0.9513	0.9843	0.9114	0.9423	0.9466	0.9211	1.0	0.9783	0.9507	0.9817	1.0
0.0239	15.0	7500	0.1638	0.9665	0.9566	0.9687	0.9606	0.9665	0.9695	0.9665	0.9671	0.9938	0.9697	0.9513	0.9869	0.9114	0.9462	0.9786	0.9306	1.0	0.9783	0.9458	1.0	1.0
0.0219	16.0	8000	0.1640	0.964	0.9542	0.9654	0.9578	0.964	0.9670	0.964	0.9646	0.9897	0.9697	0.9484	0.9869	0.9114	0.9423	0.9786	0.9180	1.0	0.9783	0.9458	0.9817	1.0
0.0204	17.0	8500	0.1653	0.9657	0.9560	0.9676	0.9600	0.9657	0.9683	0.9657	0.9662	0.9938	0.9663	0.9513	0.9869	0.9114	0.9462	0.9786	0.9338	1.0	0.9783	0.9409	0.9908	1.0
0.0187	18.0	9000	0.1824	0.9647	0.9552	0.9661	0.9590	0.9647	0.9671	0.9647	0.9652	0.9938	0.9646	0.9513	0.9869	0.9114	0.9385	0.9786	0.9338	1.0	0.9783	0.9409	0.9817	1.0
0.0165	19.0	9500	0.1762	0.9653	0.9558	0.9660	0.9587	0.9653	0.9686	0.9653	0.9659	0.9917	0.9764	0.9542	0.9869	0.9114	0.9423	0.9715	0.9211	1.0	0.9746	0.9458	0.9817	1.0
0.0147	20.0	10000	0.1952	0.9653	0.9578	0.9664	0.9610	0.9653	0.9669	0.9653	0.9655	0.9917	0.9680	0.9542	0.9869	0.9141	0.9385	0.9786	0.9338	1.0	0.9746	0.9409	0.9817	1.0
0.0134	21.0	10500	0.2101	0.964	0.9556	0.9648	0.9588	0.964	0.9660	0.964	0.9643	0.9917	0.9680	0.9570	0.9869	0.9114	0.9385	0.9680	0.9338	1.0	0.9746	0.9310	0.9817	1.0
0.0117	22.0	11000	0.1971	0.9647	0.9560	0.9656	0.9593	0.9647	0.9669	0.9647	0.9651	0.9938	0.9697	0.9513	0.9843	0.9114	0.9423	0.9786	0.9338	1.0	0.9746	0.9310	0.9817	1.0
0.0113	23.0	11500	0.2037	0.9645	0.9545	0.9655	0.9579	0.9645	0.9677	0.9645	0.9652	0.9917	0.9697	0.9542	0.9869	0.9114	0.9385	0.9715	0.9338	1.0	0.9710	0.9409	0.9817	1.0
0.0108	24.0	12000	0.2033	0.9645	0.9542	0.9657	0.9579	0.9645	0.9677	0.9645	0.9652	0.9917	0.9663	0.9542	0.9869	0.9114	0.9385	0.9715	0.9369	1.0	0.9746	0.9409	0.9817	1.0
0.0098	25.0	12500	0.1980	0.9647	0.9555	0.9655	0.9585	0.9647	0.9678	0.9647	0.9654	0.9917	0.9731	0.9484	0.9843	0.9114	0.9538	0.9715	0.9338	1.0	0.9710	0.9310	0.9817	1.0
0.0096	26.0	13000	0.2104	0.9627	0.9534	0.9642	0.9567	0.9627	0.9659	0.9627	0.9634	0.9938	0.9613	0.9484	0.9816	0.9114	0.9385	0.9751	0.9369	1.0	0.9746	0.9310	0.9817	1.0
0.0083	27.0	13500	0.2138	0.9645	0.9548	0.9665	0.9582	0.9645	0.9681	0.9645	0.9652	0.9917	0.9680	0.9542	0.9869	0.9114	0.9385	0.9680	0.9338	1.0	0.9710	0.9409	1.0	1.0
0.0084	28.0	14000	0.2180	0.9633	0.9542	0.9646	0.9578	0.9633	0.9657	0.9633	0.9637	0.9917	0.9646	0.9484	0.9869	0.9114	0.9385	0.9680	0.9369	1.0	0.9710	0.9409	0.9817	1.0
0.0076	29.0	14500	0.2196	0.9623	0.9529	0.9635	0.9565	0.9623	0.9648	0.9623	0.9627	0.9917	0.9630	0.9398	0.9869	0.9141	0.9385	0.9715	0.9369	1.0	0.9710	0.9310	0.9817	1.0
0.006	30.0	15000	0.2320	0.961	0.9511	0.9626	0.9545	0.961	0.9645	0.961	0.9617	0.9917	0.9663	0.9226	0.9869	0.9114	0.9385	0.9715	0.9338	1.0	0.9746	0.9261	0.9908	1.0
0.0067	31.0	15500	0.2285	0.9627	0.9527	0.9654	0.9567	0.9627	0.9663	0.9627	0.9634	0.9917	0.9630	0.9284	0.9869	0.9114	0.9385	0.9786	0.9369	1.0	0.9746	0.9409	1.0	1.0
0.0068	32.0	16000	0.2252	0.9625	0.9530	0.9640	0.9566	0.9625	0.9652	0.9625	0.9630	0.9917	0.9646	0.9370	0.9869	0.9141	0.9385	0.9715	0.9338	1.0	0.9710	0.9409	0.9817	1.0
0.0062	33.0	16500	0.2291	0.963	0.9530	0.9641	0.9563	0.963	0.9664	0.963	0.9637	0.9917	0.9663	0.9427	0.9869	0.9114	0.9385	0.9786	0.9338	1.0	0.9710	0.9310	0.9817	1.0
0.005	34.0	17000	0.2376	0.9627	0.9529	0.9641	0.9563	0.9627	0.9660	0.9627	0.9634	0.9917	0.9630	0.9456	0.9869	0.9114	0.9385	0.9751	0.9369	1.0	0.9710	0.9310	0.9817	1.0
0.0055	35.0	17500	0.2371	0.9627	0.9519	0.9640	0.9557	0.9627	0.9662	0.9627	0.9635	0.9917	0.9646	0.9398	0.9869	0.9114	0.9385	0.9715	0.9369	1.0	0.9783	0.9310	0.9817	1.0
0.0059	36.0	18000	0.2300	0.9623	0.9532	0.9633	0.9562	0.9623	0.9653	0.9623	0.9628	0.9917	0.9663	0.9427	0.9869	0.9114	0.9423	0.9751	0.9274	1.0	0.9710	0.9261	0.9817	1.0
0.0045	37.0	18500	0.2400	0.9625	0.9525	0.9639	0.9561	0.9625	0.9656	0.9625	0.9631	0.9917	0.9630	0.9370	0.9869	0.9114	0.9423	0.9786	0.9369	1.0	0.9746	0.9261	0.9817	1.0
0.0042	38.0	19000	0.2514	0.9627	0.9532	0.9638	0.9564	0.9627	0.9660	0.9627	0.9634	0.9917	0.9646	0.9456	0.9869	0.9114	0.9385	0.9751	0.9338	1.0	0.9746	0.9261	0.9817	1.0
0.0044	39.0	19500	0.2389	0.9635	0.9533	0.9646	0.9570	0.9635	0.9665	0.9635	0.9641	0.9917	0.9646	0.9513	0.9869	0.9114	0.9385	0.9715	0.9369	1.0	0.9746	0.9310	0.9817	1.0
0.0045	40.0	20000	0.2412	0.9613	0.9509	0.9625	0.9544	0.9613	0.9647	0.9613	0.9619	0.9917	0.9663	0.9226	0.9869	0.9114	0.9423	0.9751	0.9306	1.0	0.9783	0.9261	0.9817	1.0
0.0039	41.0	20500	0.2447	0.9617	0.9516	0.9629	0.9550	0.9617	0.9652	0.9617	0.9625	0.9917	0.9663	0.9370	0.9869	0.9114	0.9423	0.9715	0.9274	1.0	0.9710	0.9310	0.9817	1.0
0.0033	42.0	21000	0.2442	0.962	0.9515	0.9633	0.9551	0.962	0.9655	0.962	0.9627	0.9917	0.9646	0.9370	0.9869	0.9114	0.9385	0.9751	0.9306	1.0	0.9746	0.9310	0.9817	1.0
0.0042	43.0	21500	0.2431	0.963	0.9527	0.9640	0.9561	0.963	0.9665	0.963	0.9637	0.9917	0.9663	0.9484	0.9869	0.9114	0.9385	0.9751	0.9306	1.0	0.9710	0.9310	0.9817	1.0
0.0031	44.0	22000	0.2500	0.9625	0.9521	0.9636	0.9556	0.9625	0.9661	0.9625	0.9632	0.9917	0.9646	0.9398	0.9869	0.9114	0.9385	0.9751	0.9369	1.0	0.9746	0.9261	0.9817	1.0
0.004	45.0	22500	0.2495	0.9627	0.9520	0.9640	0.9557	0.9627	0.9663	0.9627	0.9635	0.9917	0.9646	0.9398	0.9869	0.9114	0.9385	0.9751	0.9369	1.0	0.9746	0.9310	0.9817	1.0
0.0028	46.0	23000	0.2459	0.9627	0.9525	0.9639	0.9561	0.9627	0.9660	0.9627	0.9634	0.9917	0.9646	0.9398	0.9869	0.9114	0.9423	0.9751	0.9369	1.0	0.9746	0.9261	0.9817	1.0
0.0028	47.0	23500	0.2476	0.963	0.9527	0.9640	0.9562	0.963	0.9663	0.963	0.9637	0.9917	0.9646	0.9484	0.9869	0.9114	0.9385	0.9715	0.9369	1.0	0.9746	0.9261	0.9817	1.0
0.003	48.0	24000	0.2437	0.9625	0.9524	0.9641	0.9559	0.9625	0.9659	0.9625	0.9632	0.9917	0.9663	0.9370	0.9869	0.9114	0.9423	0.9751	0.9306	1.0	0.9746	0.9261	0.9908	1.0
0.0034	49.0	24500	0.2444	0.9623	0.9524	0.9632	0.9556	0.9623	0.9655	0.9623	0.9629	0.9917	0.9680	0.9398	0.9869	0.9114	0.9423	0.9751	0.9243	1.0	0.9746	0.9261	0.9817	1.0
0.0027	50.0	25000	0.2447	0.9625	0.9524	0.9635	0.9558	0.9625	0.9657	0.9625	0.9631	0.9917	0.9663	0.9427	0.9869	0.9114	0.9423	0.9715	0.9306	1.0	0.9746	0.9261	0.9817	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-v4-50

bert-base-multilingual-cased-2-contract-sections-classification-v4-50

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

marcelovidigal/
bert-base-multilingual-cased-2-contract-sections-classification-v4-50