bert-base-portuguese-cased-finetuned-RM-4

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: 3.5131

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss
No log	1.0	3	8.0029
No log	2.0	6	7.4474
No log	3.0	9	7.0332
No log	4.0	12	6.5949
No log	5.0	15	6.2000
No log	6.0	18	5.8132
No log	7.0	21	5.4553
No log	8.0	24	5.1585
No log	9.0	27	4.8569
No log	10.0	30	4.6165
No log	11.0	33	4.4028
No log	12.0	36	4.2049
No log	13.0	39	4.0424
No log	14.0	42	3.9082
No log	15.0	45	3.7785
No log	16.0	48	3.6775
No log	17.0	51	3.6074
No log	18.0	54	3.5587
No log	19.0	57	3.5266
No log	20.0	60	3.5131

Framework versions

• Transformers 4.37.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.15.2