ellis-v1-emotion-negative-emotions

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: 0.8542
• Accuracy: 0.7383

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: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9602	1.0	7088	0.9456	0.7097
0.8215	2.0	14176	0.8766	0.7275
0.7029	3.0	21264	0.8542	0.7383
0.649	4.0	28352	0.8720	0.7409
0.5563	5.0	35440	0.8653	0.7475
0.4794	6.0	42528	0.9512	0.7448
0.4185	7.0	49616	1.0037	0.7444
0.3406	8.0	56704	1.1219	0.7419
0.2797	9.0	63792	1.2022	0.7414
0.2454	10.0	70880	1.2883	0.7416
0.2024	11.0	77968	1.5484	0.7406
0.1678	12.0	85056	1.6934	0.7383
0.1514	13.0	92144	1.8360	0.7405
0.1412	14.0	99232	2.0350	0.7413
0.1157	15.0	106320	2.1713	0.7382
0.1121	16.0	113408	2.4204	0.7356
0.1143	17.0	120496	2.4742	0.7382
0.102	18.0	127584	2.6642	0.7393
0.0855	19.0	134672	2.7971	0.7383
0.0728	20.0	141760	2.8854	0.7408
0.0731	21.0	148848	2.9908	0.7385
0.057	22.0	155936	3.1398	0.7375
0.0685	23.0	163024	3.1988	0.7387
0.0436	24.0	170112	3.3513	0.7382
0.049	25.0	177200	3.4022	0.7355
0.0425	26.0	184288	3.4472	0.7375
0.0436	27.0	191376	3.5385	0.7410
0.0394	28.0	198464	3.5313	0.7406
0.0263	29.0	205552	3.5983	0.7435
0.0237	30.0	212640	3.6763	0.7396
0.0399	31.0	219728	3.6650	0.7417
0.0195	32.0	226816	3.8180	0.7413
0.0176	33.0	233904	3.8228	0.7427
0.0151	34.0	240992	3.8721	0.7432
0.0148	35.0	248080	3.9694	0.7394
0.0201	36.0	255168	3.9369	0.7427
0.0179	37.0	262256	4.0059	0.7407
0.0201	38.0	269344	4.0857	0.7436
0.005	39.0	276432	4.0892	0.7428
0.0083	40.0	283520	4.1299	0.7430
0.0071	41.0	290608	4.1631	0.7452
0.0088	42.0	297696	4.1157	0.7456
0.0156	43.0	304784	4.1467	0.7452
0.0086	44.0	311872	4.1959	0.7448
0.0052	45.0	318960	4.1698	0.7443
0.0031	46.0	326048	4.2124	0.7437
0.001	47.0	333136	4.2123	0.7460
0.0018	48.0	340224	4.2051	0.7459
0.0032	49.0	347312	4.1769	0.7451
0.002	50.0	354400	4.1918	0.7451

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0