Neuria_BERT_CONTEXTO

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: 0.1801
• Accuracy: 0.9190
• Precision Micro: 0.9539
• Recall Micro: 0.9462
• F1 Micro: 0.9501
• F1 Macro: 0.9376

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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision Micro	Recall Micro	F1 Micro	F1 Macro
1.1891	1.0	81	0.9581	0.1558	0.4034	0.8360	0.5442	0.4684
0.6921	2.0	162	0.4294	0.5545	0.6583	0.9113	0.7644	0.6614
0.3172	3.0	243	0.2589	0.8442	0.8599	0.9570	0.9059	0.8296
0.1757	4.0	324	0.2079	0.9003	0.9267	0.9516	0.9390	0.9315
0.1175	5.0	405	0.1720	0.9097	0.9489	0.9489	0.9489	0.9493
0.0909	6.0	486	0.1789	0.9190	0.9515	0.9489	0.9502	0.9509
0.073	7.0	567	0.1692	0.9283	0.9543	0.9543	0.9543	0.9399
0.064	8.0	648	0.1829	0.9221	0.9542	0.9516	0.9529	0.9398
0.0585	9.0	729	0.1869	0.9221	0.9568	0.9516	0.9542	0.9499
0.0562	10.0	810	0.1801	0.9190	0.9539	0.9462	0.9501	0.9376

Framework versions

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