belisards/mmnlab

final

This model is a fine-tuned version of neuralmind/bert-large-portuguese-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.3507
• Accuracy: 0.8945
• F1: 0.8863
• Recall: 0.8760
• Precision: 0.8968

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: 64
• eval_batch_size: 64
• seed: 5151
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 100
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
0.485	0.9756	80	0.3916	0.8182	0.7984	0.7674	0.8319
0.3395	1.9512	160	0.3039	0.8764	0.8547	0.7752	0.9524
0.2139	2.9268	240	0.3122	0.8691	0.8548	0.8217	0.8908
0.084	3.9024	320	0.3507	0.8945	0.8863	0.8760	0.8968
0.058	4.8780	400	0.5087	0.8727	0.8571	0.8140	0.9052
0.0389	5.8537	480	0.4579	0.8982	0.888	0.8605	0.9174
0.0264	6.8293	560	0.5052	0.8873	0.8765	0.8527	0.9016

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1