NOTA: Este modelo será utilizado para comparação com a Versão 2.0. Poderá ser excluido após esta validação/comparação

ellis-v3-emotion-leadership

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.0202
• Accuracy: 0.8402

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 70

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6209	1.0	1479	0.5094	0.8014
0.4893	2.0	2958	0.4642	0.8166
0.4244	3.0	4437	0.4640	0.8284
0.3607	4.0	5916	0.4596	0.8333
0.3021	5.0	7395	0.4892	0.8322
0.2391	6.0	8874	0.5455	0.8288
0.2066	7.0	10353	0.6553	0.8288
0.1669	8.0	11832	0.6856	0.8387
0.1515	9.0	13311	0.8654	0.8280
0.1154	10.0	14790	0.8985	0.8322
0.094	11.0	16269	1.1159	0.8280
0.0822	12.0	17748	1.1954	0.8227
0.082	13.0	19227	1.2213	0.8341
0.0656	14.0	20706	1.2533	0.8375
0.0592	15.0	22185	1.3723	0.8284
0.0539	16.0	23664	1.4376	0.8326
0.0533	17.0	25143	1.4746	0.8291
0.044	18.0	26622	1.4234	0.8288
0.0411	19.0	28101	1.4971	0.8253
0.0343	20.0	29580	1.5132	0.8284
0.0359	21.0	31059	1.5020	0.8360
0.0309	22.0	32538	1.6418	0.8356
0.0416	23.0	34017	1.4984	0.8303
0.0314	24.0	35496	1.5713	0.8341
0.0316	25.0	36975	1.5679	0.8352
0.0281	26.0	38454	1.6399	0.8311
0.0179	27.0	39933	1.7032	0.8231
0.0326	28.0	41412	1.6551	0.8330
0.0178	29.0	42891	1.7136	0.8284
0.0149	30.0	44370	1.7317	0.8288
0.0211	31.0	45849	1.6790	0.8314
0.0221	32.0	47328	1.7909	0.8280
0.0179	33.0	48807	1.8027	0.8314
0.022	34.0	50286	1.7754	0.8299
0.0198	35.0	51765	1.7498	0.8295
0.0124	36.0	53244	1.8098	0.8356
0.0123	37.0	54723	1.8535	0.8261
0.0103	38.0	56202	1.8827	0.8345
0.0145	39.0	57681	1.8882	0.8303
0.0162	40.0	59160	1.8174	0.8326
0.0103	41.0	60639	1.8350	0.8368
0.0103	42.0	62118	1.7853	0.8390
0.0136	43.0	63597	1.7032	0.8368
0.0099	44.0	65076	1.8274	0.8318
0.0074	45.0	66555	1.8598	0.8333
0.0108	46.0	68034	1.7978	0.8413
0.0063	47.0	69513	1.8116	0.8364
0.0112	48.0	70992	1.8066	0.8356
0.0038	49.0	72471	1.9092	0.8352
0.005	50.0	73950	1.9159	0.8356
0.0035	51.0	75429	1.8669	0.8379
0.0067	52.0	76908	1.9222	0.8333
0.0049	53.0	78387	1.8417	0.8398
0.0034	54.0	79866	2.0452	0.8311
0.0056	55.0	81345	1.9375	0.8349
0.0014	56.0	82824	1.9941	0.8322
0.0004	57.0	84303	2.0133	0.8349
0.0017	58.0	85782	2.0038	0.8356
0.0009	59.0	87261	2.0347	0.8356
0.0015	60.0	88740	1.9901	0.8368
0.0014	61.0	90219	2.0233	0.8368
0.0022	62.0	91698	2.0148	0.8356
0.0012	63.0	93177	1.9823	0.8383
0.0014	64.0	94656	2.0099	0.8368
0.0034	65.0	96135	1.9925	0.8402
0.0009	66.0	97614	2.0088	0.8390
0.0006	67.0	99093	2.0141	0.8394
0.0	68.0	100572	2.0199	0.8417
0.0012	69.0	102051	2.0187	0.8394
0.0	70.0	103530	2.0202	0.8402

Framework versions

• Transformers 4.39.3
• Pytorch 2.1.0
• Datasets 2.18.0
• Tokenizers 0.15.2