not-ner-v1

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

• Loss: 0.1680
• Accuracy: 0.9337
• Precision: 0.9334
• Recall: 0.9337
• F1: 0.9333

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: 20
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.3465	0.0799	200	0.3024	0.8719	0.8787	0.8719	0.8737
0.2925	0.1599	400	0.2530	0.9045	0.9039	0.9045	0.9041
0.2362	0.2398	600	0.2383	0.9089	0.9084	0.9089	0.9085
0.239	0.3197	800	0.2083	0.9169	0.9163	0.9169	0.9163
0.2149	0.3997	1000	0.2640	0.9130	0.9150	0.9130	0.9109
0.2171	0.4796	1200	0.1932	0.9211	0.9214	0.9211	0.9212
0.2056	0.5596	1400	0.1962	0.9237	0.9243	0.9237	0.9224
0.1973	0.6395	1600	0.1906	0.9258	0.9255	0.9258	0.9256
0.1912	0.7194	1800	0.1870	0.9277	0.9275	0.9277	0.9270
0.183	0.7994	2000	0.1727	0.9318	0.9317	0.9318	0.9318
0.1672	0.8793	2200	0.1809	0.9320	0.9318	0.9320	0.9313
0.1643	0.9592	2400	0.1680	0.9337	0.9334	0.9337	0.9333

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1