Instructions for reproducing the experients from the OOV anonymization paper

bayesgroup / code_transformers

Empirical Study of Transformers for Source Code & A Simple Approach for Handling Out-of-Vocabulary Identifiers in Deep Learning for Source Code

Other

61 stars 18 forks source link

Instructions for reproducing the experients from the OOV anonymization paper #1

Open aboutzack opened 3 years ago

aboutzack commented 3 years ago

Greetings! Congrats for the great work! My question is when you will post the instructions for reproducing the experients from the OOV anonymization paper. I believe that will be of great help to those who follow this work. It is a very interesting work and I think that it disearves more attention. Thanks in advance!

serjtroshin commented 3 years ago

Hi! Thank you for reaching us. If you are interested in the implementation of OOV anonymization, it is already available. For the vm_fn tasks, please consider --anonymize order flag, and for cc task, --use_anonymized flag. As for the scripts reproducing experiments from the paper, we will provide them soon.