IuryCavalcante/TechnicalDebt-contex-classifier

results

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

• Loss: 0.9699
• Accuracy: 0.8696
• F1 Macro: 0.7453

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: 8
• 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: 12

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Macro
No log	1.0	17	1.3024	0.6522	0.4824
No log	2.0	34	1.1881	0.7826	0.6572
No log	3.0	51	0.9699	0.8696	0.7453
No log	4.0	68	0.8733	0.8261	0.7017
No log	5.0	85	0.7908	0.8261	0.7017
No log	6.0	102	0.8341	0.8261	0.7017
No log	7.0	119	0.7325	0.8261	0.7017
No log	8.0	136	0.7554	0.8261	0.7017
No log	9.0	153	0.8333	0.8261	0.6918
No log	10.0	170	0.8074	0.8261	0.6831
No log	11.0	187	0.8021	0.8261	0.6732
No log	12.0	204	0.8205	0.8261	0.6732

Framework versions

• Transformers 4.54.0
• Pytorch 2.7.1+cpu
• Datasets 4.0.0
• Tokenizers 0.21.2