RQ1: Is Neural Machine Translation a viable approach to
learn how to fix code?
Results for models' performances:
- Overview
- Models' Performances - Values
- Hyperparameter Search Configurations
Overview
Overview
Overview of the model performances, in terms of perfect fixes, for small and medium methods.
Models' Performances - Values
Models' Performances - Values
The table shows the numbers and percentages of perfect fixes generated by the models with variable beam size. The one highlighted in green are experiments with larger beam sizes not reported in the paper.
Hyperparameter Search Configurations
Hyperparameter Search Configurations
The table shows the ten different configurations of the hyperparameters tested. The one highlighted in green is the configuration achieving the best results on the validation set