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

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

• Loss: 0.2691
• Accuracy Evaluate: 0.9623
• Precision Evaluate: 0.9580
• Recall Evaluate: 0.9662
• F1 Evaluate: 0.9614
• Accuracy Sklearn: 0.9623
• Precision Sklearn: 0.9634
• Recall Sklearn: 0.9623
• F1 Sklearn: 0.9624
• Acuracia Rotulo Objeto: 0.9835
• Acuracia Rotulo Obrigacoes: 0.9461
• Acuracia Rotulo Valor: 0.9513
• Acuracia Rotulo Vigencia: 0.9948
• Acuracia Rotulo Rescisao: 0.9114
• Acuracia Rotulo Foro: 1.0
• Acuracia Rotulo Reajuste: 0.9573
• Acuracia Rotulo Fiscalizacao: 0.9274
• Acuracia Rotulo Publicacao: 1.0
• Acuracia Rotulo Pagamento: 0.9565
• Acuracia Rotulo Casos Omissos: 0.9409
• 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: 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
1.9398	1.0	1000	1.7897	0.5843	0.6911	0.5394	0.5537	0.5843	0.6859	0.5843	0.5715	0.9194	0.7155	0.5731	0.5906	0.6787	0.2038	0.7758	0.2177	0.8473	0.2210	0.9113	0.3578	0.0
0.9918	2.0	2000	0.9569	0.8448	0.8604	0.8426	0.8441	0.8448	0.8560	0.8448	0.8437	0.9545	0.7593	0.9312	0.8346	0.9114	0.9385	0.9217	0.7634	0.9310	0.6196	0.9163	0.8899	0.5824
0.5019	3.0	3000	0.5333	0.9058	0.9079	0.9179	0.9101	0.9058	0.9123	0.9058	0.9055	0.9773	0.7559	0.9341	0.9659	0.9114	0.9385	0.9751	0.8549	0.9655	0.8188	0.9163	0.9358	0.9835
0.2823	4.0	4000	0.3519	0.9237	0.9256	0.9358	0.9280	0.9237	0.9291	0.9237	0.9235	0.9835	0.7761	0.9398	0.9843	0.9252	0.9962	0.9929	0.8864	0.9754	0.8442	0.9163	0.9450	1.0
0.173	5.0	5000	0.2748	0.935	0.9328	0.9440	0.9363	0.935	0.9384	0.935	0.9352	0.9690	0.8485	0.9542	0.9843	0.9169	1.0	0.9964	0.8675	0.9901	0.8659	0.9163	0.9633	1.0
0.1141	6.0	6000	0.2648	0.9337	0.9327	0.9458	0.9369	0.9337	0.9388	0.9337	0.9342	0.9752	0.8064	0.9456	0.9948	0.9197	1.0	0.9964	0.8833	0.9951	0.8949	0.9212	0.9633	1.0
0.0959	7.0	7000	0.2332	0.9447	0.9412	0.9512	0.9444	0.9447	0.9473	0.9447	0.9450	0.9669	0.8956	0.9513	0.9948	0.9141	1.0	0.9964	0.8675	0.9901	0.8949	0.9212	0.9725	1.0
0.0785	8.0	8000	0.2453	0.94	0.9371	0.9496	0.9412	0.94	0.9440	0.94	0.9403	0.9773	0.8468	0.9398	0.9948	0.9141	1.0	1.0	0.8927	0.9901	0.8949	0.9212	0.9725	1.0
0.0586	9.0	9000	0.2317	0.945	0.9407	0.9522	0.9445	0.945	0.9481	0.945	0.9453	0.9752	0.8822	0.9427	0.9948	0.9086	1.0	1.0	0.8927	0.9901	0.8986	0.9212	0.9725	1.0
0.049	10.0	10000	0.2366	0.9447	0.9399	0.9528	0.9443	0.9447	0.9482	0.9447	0.9451	0.9752	0.8771	0.9427	0.9974	0.9086	1.0	1.0	0.8927	0.9901	0.8949	0.9261	0.9817	1.0
0.0506	11.0	11000	0.2213	0.9515	0.9456	0.9564	0.9493	0.9515	0.9538	0.9515	0.9517	0.9814	0.9209	0.9312	0.9974	0.9141	1.0	1.0	0.8896	0.9901	0.8913	0.9360	0.9817	1.0
0.0362	12.0	12000	0.2341	0.9515	0.9453	0.9588	0.9503	0.9515	0.9540	0.9515	0.9518	0.9731	0.9074	0.9312	0.9974	0.9086	1.0	0.9964	0.8896	0.9951	0.9384	0.9360	0.9908	1.0
0.0412	13.0	13000	0.2332	0.9487	0.9444	0.9555	0.9485	0.9487	0.9510	0.9487	0.9489	0.9814	0.8956	0.9341	0.9974	0.9197	1.0	0.9929	0.8927	0.9951	0.8949	0.9360	0.9817	1.0
0.0261	14.0	14000	0.2384	0.956	0.9497	0.9605	0.9533	0.956	0.9583	0.956	0.9563	0.9814	0.9360	0.9398	0.9974	0.9086	1.0	0.9822	0.8864	0.9951	0.9420	0.9261	0.9908	1.0
0.0268	15.0	15000	0.2328	0.9583	0.9516	0.9636	0.9558	0.9583	0.9606	0.9583	0.9585	0.9773	0.9360	0.9341	0.9974	0.9086	1.0	0.9964	0.8896	1.0	0.9601	0.9360	0.9908	1.0
0.0248	16.0	16000	0.2271	0.9593	0.9528	0.9641	0.9569	0.9593	0.9613	0.9593	0.9595	0.9814	0.9394	0.9370	0.9974	0.9086	1.0	0.9929	0.8896	1.0	0.9601	0.9360	0.9908	1.0
0.0235	17.0	17000	0.2273	0.9607	0.9542	0.9651	0.9580	0.9607	0.9629	0.9607	0.9610	0.9773	0.9495	0.9427	0.9974	0.9086	1.0	0.9929	0.8896	1.0	0.9565	0.9409	0.9908	1.0
0.0214	18.0	18000	0.2160	0.9607	0.9562	0.9655	0.9599	0.9607	0.9622	0.9607	0.9609	0.9876	0.9327	0.9456	0.9974	0.9141	1.0	0.9822	0.9117	0.9951	0.9529	0.9409	0.9908	1.0
0.0207	19.0	19000	0.2345	0.9597	0.9534	0.9644	0.9575	0.9597	0.9618	0.9597	0.9600	0.9876	0.9327	0.9484	0.9974	0.9058	1.0	0.9644	0.9148	1.0	0.9638	0.9310	0.9908	1.0
0.0166	20.0	20000	0.2340	0.9587	0.9538	0.9635	0.9576	0.9587	0.9602	0.9587	0.9589	0.9814	0.9360	0.9484	0.9974	0.9086	1.0	0.9573	0.9085	1.0	0.9565	0.9409	0.9908	1.0
0.0167	21.0	21000	0.2487	0.96	0.9532	0.9639	0.9570	0.96	0.9621	0.96	0.9603	0.9814	0.9495	0.9484	0.9974	0.9086	1.0	0.9680	0.8896	1.0	0.9565	0.9409	0.9908	1.0
0.0094	22.0	22000	0.2476	0.9593	0.9535	0.9633	0.9571	0.9593	0.9610	0.9593	0.9595	0.9835	0.9444	0.9484	0.9974	0.9086	1.0	0.9680	0.8896	1.0	0.9565	0.9360	0.9908	1.0
0.0119	23.0	23000	0.2461	0.9597	0.9545	0.9650	0.9587	0.9597	0.9614	0.9597	0.9600	0.9793	0.9310	0.9484	0.9974	0.9086	1.0	0.9644	0.9274	1.0	0.9565	0.9409	0.9908	1.0
0.0148	24.0	24000	0.2768	0.9567	0.9510	0.9604	0.9539	0.9567	0.9593	0.9567	0.9571	0.9814	0.9512	0.9484	0.9921	0.9058	1.0	0.9644	0.8801	1.0	0.9348	0.9360	0.9908	1.0
0.0143	25.0	25000	0.2564	0.9587	0.9531	0.9630	0.9567	0.9587	0.9604	0.9587	0.9589	0.9814	0.9444	0.9484	0.9974	0.9058	1.0	0.9644	0.8896	1.0	0.9601	0.9360	0.9908	1.0
0.0088	26.0	26000	0.2479	0.9607	0.9547	0.9654	0.9589	0.9607	0.9624	0.9607	0.9610	0.9814	0.9394	0.9456	0.9974	0.9058	1.0	0.9680	0.9243	1.0	0.9565	0.9409	0.9908	1.0
0.01	27.0	27000	0.2526	0.9595	0.9541	0.9637	0.9581	0.9595	0.9611	0.9595	0.9597	0.9814	0.9310	0.9513	0.9974	0.9058	1.0	0.9609	0.9274	1.0	0.9601	0.9409	0.9725	1.0
0.0091	28.0	28000	0.2517	0.9613	0.9561	0.9659	0.9601	0.9613	0.9628	0.9613	0.9615	0.9814	0.9377	0.9513	0.9974	0.9058	1.0	0.9644	0.9274	1.0	0.9601	0.9409	0.9908	1.0
0.0113	29.0	29000	0.2497	0.9607	0.9565	0.9647	0.9598	0.9607	0.9617	0.9607	0.9608	0.9814	0.9478	0.9484	0.9974	0.9114	1.0	0.9644	0.9054	1.0	0.9529	0.9409	0.9908	1.0
0.0104	30.0	30000	0.2702	0.9597	0.9547	0.9625	0.9574	0.9597	0.9613	0.9597	0.9599	0.9835	0.9545	0.9484	0.9948	0.9058	1.0	0.9644	0.8927	1.0	0.9601	0.9261	0.9817	1.0
0.0114	31.0	31000	0.2568	0.9613	0.9575	0.9654	0.9607	0.9613	0.9623	0.9613	0.9614	0.9793	0.9461	0.9484	0.9974	0.9114	1.0	0.9644	0.9148	1.0	0.9565	0.9409	0.9908	1.0
0.009	32.0	32000	0.2590	0.9597	0.9547	0.9641	0.9584	0.9597	0.9611	0.9597	0.9599	0.9752	0.9478	0.9484	0.9948	0.9086	1.0	0.9644	0.9022	1.0	0.9601	0.9409	0.9908	1.0
0.0106	33.0	33000	0.2540	0.9613	0.9575	0.9640	0.9602	0.9613	0.9622	0.9613	0.9614	0.9835	0.9512	0.9484	0.9921	0.9114	1.0	0.9644	0.9117	1.0	0.9565	0.9409	0.9725	1.0
0.0083	34.0	34000	0.2545	0.962	0.9579	0.9656	0.9611	0.962	0.9631	0.962	0.9621	0.9793	0.9478	0.9513	0.9948	0.9114	1.0	0.9644	0.9243	1.0	0.9565	0.9409	0.9817	1.0
0.008	35.0	35000	0.2586	0.9617	0.9559	0.9654	0.9597	0.9617	0.9632	0.9617	0.9620	0.9814	0.9512	0.9513	0.9948	0.9058	1.0	0.9537	0.9243	1.0	0.9565	0.9409	0.9908	1.0
0.0042	36.0	36000	0.2533	0.9623	0.9581	0.9653	0.9611	0.9623	0.9633	0.9623	0.9624	0.9855	0.9495	0.9513	0.9948	0.9114	1.0	0.9537	0.9243	1.0	0.9565	0.9409	0.9817	1.0
0.0056	37.0	37000	0.2637	0.9615	0.9565	0.9650	0.9598	0.9615	0.9628	0.9615	0.9616	0.9876	0.9478	0.9513	0.9921	0.9058	1.0	0.9537	0.9243	1.0	0.9601	0.9310	0.9908	1.0
0.0075	38.0	38000	0.2632	0.9613	0.9573	0.9645	0.9603	0.9613	0.9623	0.9613	0.9614	0.9835	0.9461	0.9513	0.9948	0.9114	1.0	0.9502	0.9274	1.0	0.9565	0.9360	0.9817	1.0
0.0041	39.0	39000	0.2641	0.9595	0.9557	0.9629	0.9586	0.9595	0.9605	0.9595	0.9596	0.9752	0.9529	0.9513	0.9948	0.9114	1.0	0.9573	0.8959	1.0	0.9565	0.9409	0.9817	1.0
0.006	40.0	40000	0.2645	0.962	0.9574	0.9654	0.9607	0.962	0.9631	0.962	0.9621	0.9855	0.9461	0.9513	0.9948	0.9114	1.0	0.9573	0.9243	1.0	0.9565	0.9409	0.9817	1.0
0.0064	41.0	41000	0.2618	0.9603	0.9563	0.9641	0.9594	0.9603	0.9612	0.9603	0.9603	0.9855	0.9478	0.9513	0.9921	0.9114	1.0	0.9573	0.8991	1.0	0.9565	0.9409	0.9908	1.0
0.0046	42.0	42000	0.2679	0.9605	0.9573	0.9640	0.9600	0.9605	0.9613	0.9605	0.9605	0.9835	0.9529	0.9513	0.9921	0.9114	1.0	0.9573	0.8959	1.0	0.9565	0.9409	0.9908	1.0
0.0065	43.0	43000	0.2668	0.962	0.9578	0.9653	0.9609	0.962	0.9630	0.962	0.9621	0.9835	0.9495	0.9513	0.9921	0.9114	1.0	0.9573	0.9243	1.0	0.9565	0.9409	0.9817	1.0
0.0026	44.0	44000	0.2724	0.96	0.9550	0.9635	0.9581	0.96	0.9615	0.96	0.9602	0.9835	0.9562	0.9513	0.9843	0.9058	1.0	0.9573	0.8959	1.0	0.9601	0.9409	0.9908	1.0
0.0081	45.0	45000	0.2671	0.9617	0.9577	0.9653	0.9609	0.9617	0.9628	0.9617	0.9619	0.9835	0.9444	0.9513	0.9948	0.9114	1.0	0.9573	0.9274	1.0	0.9565	0.9409	0.9817	1.0
0.005	46.0	46000	0.2672	0.962	0.9576	0.9654	0.9609	0.962	0.9631	0.962	0.9621	0.9835	0.9461	0.9513	0.9948	0.9114	1.0	0.9573	0.9274	1.0	0.9565	0.9409	0.9817	1.0
0.0048	47.0	47000	0.2694	0.9623	0.9577	0.9662	0.9612	0.9623	0.9635	0.9623	0.9624	0.9835	0.9461	0.9513	0.9948	0.9114	1.0	0.9573	0.9274	1.0	0.9565	0.9409	0.9908	1.0
0.0068	48.0	48000	0.2693	0.9627	0.9576	0.9663	0.9611	0.9627	0.9641	0.9627	0.9629	0.9835	0.9512	0.9513	0.9921	0.9114	1.0	0.9573	0.9274	1.0	0.9565	0.9409	0.9908	1.0
0.0041	49.0	49000	0.2695	0.9623	0.9580	0.9662	0.9614	0.9623	0.9634	0.9623	0.9624	0.9835	0.9461	0.9513	0.9948	0.9114	1.0	0.9573	0.9274	1.0	0.9565	0.9409	0.9908	1.0
0.0037	50.0	50000	0.2691	0.9623	0.9580	0.9662	0.9614	0.9623	0.9634	0.9623	0.9624	0.9835	0.9461	0.9513	0.9948	0.9114	1.0	0.9573	0.9274	1.0	0.9565	0.9409	0.9908	1.0

Framework versions

• Transformers 4.48.3
• Pytorch 2.6.0+cu124
• Datasets 3.3.0
• Tokenizers 0.21.0