Neuria_BERT_Graficos_2025_02_05

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: 0.3226
• Accuracy: 0.9634

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: 5e-05
• train_batch_size: 4
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• 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
1.1144	1.0	29	0.5802	0.6707
0.527	2.0	58	0.6142	0.7561
0.2708	3.0	87	0.4024	0.8780
0.0954	4.0	116	0.4247	0.9024
0.0864	5.0	145	0.3698	0.9024
0.0146	6.0	174	0.4584	0.9146
0.0213	7.0	203	0.4625	0.9268
0.0267	8.0	232	0.3833	0.9268
0.0009	9.0	261	0.2960	0.9512
0.0007	10.0	290	0.2934	0.9512
0.0005	11.0	319	0.2940	0.9634
0.0005	12.0	348	0.3021	0.9634
0.0004	13.0	377	0.3062	0.9634
0.0004	14.0	406	0.3084	0.9634
0.0004	15.0	435	0.3158	0.9634
0.0003	16.0	464	0.3173	0.9634
0.0003	17.0	493	0.3146	0.9634
0.0003	18.0	522	0.3169	0.9634
0.0003	19.0	551	0.3215	0.9634
0.0002	20.0	580	0.3226	0.9634

Framework versions

• Transformers 4.44.1
• Pytorch 2.4.1
• Datasets 2.19.1
• Tokenizers 0.19.1