Code Prediction by Feeding Trees to Transformers

• Published in: Proceedings / International Conference on Software Engineering pp. 150 - 162
• Main Authors: Kim, Seohyun, Zhao, Jinman, Tian, Yuchi, Chandra, Satish
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2021
• Subjects: autocomplete; code embedding; code prediction; Computer architecture; Pipelines; Proposals; Python; Software engineering; Syntactics

Code prediction, more specifically autocomplete, has become an essential feature in modern IDEs. Autocomplete is more effective when the desired next token is at (or close to) the top of the list of potential completions offered by the IDE at cursor position. This is where the strength of the underl...