oracle_class_bin

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: 0.1746
• Precision: 0.8254
• Recall: 0.7923
• Accuracy: 0.9615
• F1: 0.8085

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: 24
• eval_batch_size: 24
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 5
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	Accuracy	F1
0.1271	0.8407	1800	0.1045	0.7513	0.8544	0.9560	0.7996
0.0913	1.6815	3600	0.1075	0.8110	0.7968	0.9601	0.8038
0.0791	2.5222	5400	0.1283	0.8287	0.7885	0.9615	0.8081
0.0553	3.3629	7200	0.1272	0.8160	0.8067	0.9615	0.8113
0.0384	4.2036	9000	0.1746	0.8254	0.7923	0.9615	0.8085

Framework versions

• Transformers 4.41.2
• Pytorch 2.1.0
• Datasets 2.20.0
• Tokenizers 0.19.1