In this work, we present a large-scale study investigating the potential of data-driven approaches to support automated variable renaming.
-
- The code for training and testing the model is available here
-
- Under this path, we also relase extra scripts we used for the experiment. (e.g; the java typename ids resolver)
-
- CSV Files: Each file contains multiple columns: methodBody, methodTokens, methodIndex, extractedVars, extractedFrom. Some of those files, also contain extra columns mainly useful for keeping track of general statistic
- Model-Specific-Dataset: For each one of the experimented model, we created a specific (train,test and eval dataset) based on the aforementioned CSVs
-
- Here, you can find the model we trained and made it publicly available
-
- Under this folder you will find all the predictions concerning the experiment