O MELHOR SINGLE ESTRATéGIA A UTILIZAR PARA ROBERTA PIRES

Edit RoBERTa is an extension of BERT with changes to the pretraining procedure. The modifications include: training the model longer, with bigger batches, over more dataRoBERTa has almost similar architecture as compare to BERT, but in order to improve the results on BERT architecture, the authors made some simple design changes in its architecture