metadata

license: mit
base_model: neuralmind/bert-base-portuguese-cased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: nees-bert-base-portuguese-cased-finetuned-ner
    results: []

nees-bert-base-portuguese-cased-finetuned-ner

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

Loss: 0.0255
Precision: 0.6545
Recall: 0.7802
F1: 0.7119
Accuracy: 0.9952

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0344	1.0	519	0.0189	0.0	0.0	0.0	0.9949
0.0145	2.0	1038	0.0112	0.5050	0.4737	0.4888	0.9952
0.0121	3.0	1557	0.0132	0.3684	0.1300	0.1922	0.9953
0.0107	4.0	2076	0.0239	0.6366	0.7647	0.6948	0.9955
0.0056	5.0	2595	0.0151	0.6845	0.7121	0.6980	0.9950
0.0053	6.0	3114	0.0278	0.6432	0.7368	0.6869	0.9943
0.0047	7.0	3633	0.0199	0.5682	0.7740	0.6553	0.9953
0.0043	8.0	4152	0.0231	0.6429	0.7802	0.7049	0.9951
0.0022	9.0	4671	0.0255	0.6487	0.7833	0.7097	0.9955
0.0025	10.0	5190	0.0255	0.6545	0.7802	0.7119	0.9952

Framework versions

Transformers 4.37.2
Pytorch 2.1.0+cu121
Datasets 2.17.1
Tokenizers 0.15.2