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: 0.0360
Precision: 0.5404
Recall: 0.7230
F1: 0.6185
Accuracy: 0.9949

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.0352	1.0	599	0.0212	0.0154	0.0034	0.0055	0.9956
0.0125	2.0	1198	0.0184	0.4039	0.2770	0.3287	0.9958
0.0099	3.0	1797	0.0141	0.4528	0.4865	0.4691	0.9958
0.0082	4.0	2396	0.0279	0.4558	0.5743	0.5082	0.9958
0.0074	5.0	2995	0.0138	0.5153	0.6824	0.5872	0.9960
0.0034	6.0	3594	0.0193	0.4934	0.6284	0.5527	0.9953
0.0046	7.0	4193	0.0190	0.5305	0.7635	0.6260	0.9960
0.0024	8.0	4792	0.0286	0.5670	0.6858	0.6208	0.9955
0.0024	9.0	5391	0.0270	0.5889	0.5709	0.5798	0.9953
0.0034	10.0	5990	0.0339	0.5623	0.6858	0.6180	0.9953
0.0015	11.0	6589	0.0373	0.6122	0.7095	0.6573	0.9947
0.001	12.0	7188	0.0361	0.5519	0.6284	0.5877	0.9950
0.0005	13.0	7787	0.0353	0.5658	0.6824	0.6187	0.9950
0.0007	14.0	8386	0.0355	0.5556	0.7264	0.6296	0.9948
0.0003	15.0	8985	0.0360	0.5404	0.7230	0.6185	0.9949

Framework versions

Transformers 4.38.1
Pytorch 2.1.0+cu121
Datasets 2.18.0
Tokenizers 0.15.2