bert-base-spanish-wwm-cased-ner

This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3695
• Precision: 0.8640
• Recall: 0.9126
• F1: 0.8876
• Accuracy: 0.9378

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	280	0.3016	0.7779	0.8252	0.8009	0.9052
0.4113	2.0	560	0.2671	0.8150	0.8681	0.8407	0.9248
0.4113	3.0	840	0.2747	0.8181	0.8593	0.8382	0.9268
0.1179	4.0	1120	0.2875	0.8336	0.8978	0.8645	0.9312
0.1179	5.0	1400	0.3087	0.8529	0.9022	0.8769	0.9361
0.0608	6.0	1680	0.3449	0.8645	0.8978	0.8808	0.9351
0.0608	7.0	1960	0.3478	0.8539	0.8919	0.8725	0.9337
0.0306	8.0	2240	0.3495	0.8426	0.8963	0.8686	0.9337
0.0231	9.0	2520	0.3812	0.8660	0.9096	0.8873	0.9366
0.0231	10.0	2800	0.3346	0.8473	0.8963	0.8711	0.9386
0.0174	11.0	3080	0.3721	0.8583	0.9067	0.8818	0.9373
0.0174	12.0	3360	0.3778	0.8632	0.9067	0.8844	0.9371
0.014	13.0	3640	0.3733	0.8624	0.9096	0.8854	0.9366
0.014	14.0	3920	0.3709	0.8652	0.9126	0.8882	0.9398
0.013	15.0	4200	0.3695	0.8640	0.9126	0.8876	0.9378

Framework versions

• Transformers 4.41.1
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1