Neuria_BERT_Graficos

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

• Loss: 0.0062
• Accuracy: 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: 2e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 16
• total_train_batch_size: 256
• optimizer: Use 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
1.7306	0.8767	4	1.4663	0.3756
1.5419	1.8767	8	1.2706	0.5902
1.3248	2.8767	12	1.0288	0.6927
1.0823	3.8767	16	0.7660	0.8146
0.8438	4.8767	20	0.5362	0.9220
0.6149	5.8767	24	0.3765	0.9463
0.456	6.8767	28	0.2742	0.9512
0.3428	7.8767	32	0.2059	0.9610
0.2517	8.8767	36	0.1648	0.9659
0.1923	9.8767	40	0.1213	0.9756
0.1519	10.8767	44	0.1018	0.9756
0.1172	11.8767	48	0.0906	0.9756
0.0879	12.8767	52	0.0812	0.9805
0.0688	13.8767	56	0.0633	0.9854
0.0538	14.8767	60	0.0473	0.9854
0.0435	15.8767	64	0.0367	0.9902
0.0381	16.8767	68	0.0274	0.9902
0.027	17.8767	72	0.0330	0.9951
0.0236	18.8767	76	0.0346	0.9902
0.0211	19.8767	80	0.0269	0.9902
0.0201	20.8767	84	0.0234	0.9951
0.0176	21.8767	88	0.0203	0.9951
0.0151	22.8767	92	0.0189	0.9951
0.0152	23.8767	96	0.0108	1.0
0.0157	24.8767	100	0.0141	0.9951
0.0133	25.8767	104	0.0093	1.0
0.0123	26.8767	108	0.0153	0.9951
0.0123	27.8767	112	0.0096	1.0
0.0112	28.8767	116	0.0111	0.9951
0.0111	29.8767	120	0.0088	1.0
0.0114	30.8767	124	0.0137	0.9951
0.0104	31.8767	128	0.0085	1.0
0.012	32.8767	132	0.0066	1.0
0.0106	33.8767	136	0.0081	1.0
0.0103	34.8767	140	0.0083	1.0
0.0093	35.8767	144	0.0095	0.9951
0.0099	36.8767	148	0.0144	0.9951
0.0088	37.8767	152	0.0094	1.0
0.0085	38.8767	156	0.0075	1.0
0.0081	39.8767	160	0.0067	1.0
0.0082	40.8767	164	0.0060	1.0
0.0079	41.8767	168	0.0058	1.0
0.0078	42.8767	172	0.0058	1.0
0.0078	43.8767	176	0.0060	1.0
0.0076	44.8767	180	0.0061	1.0
0.0077	45.8767	184	0.0062	1.0
0.0079	46.8767	188	0.0064	1.0
0.0078	47.8767	192	0.0065	1.0
0.0081	48.8767	196	0.0062	1.0
0.0071	49.8767	200	0.0062	1.0

Framework versions

• Transformers 4.49.0
• Pytorch 2.4.1
• Datasets 2.19.1
• Tokenizers 0.21.0