This model is a fine-tuned version of neuralmind/bert-base-portuguese-cased on the lener_br dataset. It achieves the following results on the evaluation set:

Loss: 0.1931
Precision: 0.8384
Recall: 0.9092
F1: 0.8724
Accuracy: 0.9699

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: 16
seed: 42
optimizer: Use OptimizerNames.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: 10
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0601	1.0	979	0.1134	0.8575	0.8516	0.8546	0.9715
0.0345	2.0	1958	0.1402	0.7896	0.9022	0.8421	0.9657
0.0243	3.0	2937	0.1350	0.8124	0.9060	0.8566	0.9696
0.0256	4.0	3916	0.1592	0.7624	0.9073	0.8286	0.9640
0.0143	5.0	4895	0.1951	0.8462	0.8983	0.8715	0.9678
0.0139	6.0	5874	0.1874	0.8252	0.9110	0.8660	0.9679
0.0051	7.0	6853	0.1685	0.8301	0.9049	0.8659	0.9692
0.0067	8.0	7832	0.1931	0.8384	0.9092	0.8724	0.9699
0.0018	9.0	8811	0.2004	0.8206	0.9110	0.8634	0.9692
0.0044	10.0	9790	0.2000	0.8295	0.9090	0.8674	0.9694

Framework versions

Transformers 4.51.3
Pytorch 2.6.0+cu124
Datasets 3.5.1
Tokenizers 0.21.1

Palu1006/ner-bert-lenerbr-v2

ner-bert-lenerbr-v2

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Palu1006/
ner-bert-lenerbr-v2