TL;DR
This paper presents a transformer-based machine learning model that predicts optimal moments to invoke code completion, reducing interruptions and operational costs by intelligently filtering suggestions based on code context and developer telemetry.
Contribution
The study introduces a novel transformer-based approach for predicting code completion invocation timing, incorporating telemetry data and demonstrating practical deployment in real-world developer environments.
Findings
The transformer model outperforms baseline in prediction accuracy.
Incorporating telemetry data improves model performance.
Deployment with 34 developers shows practical effectiveness.
Abstract
Transformer-based language models are highly effective for code completion, with much research dedicated to enhancing the content of these completions. Despite their effectiveness, these models come with high operational costs and can be intrusive, especially when they suggest too often and interrupt developers who are concentrating on their work. Current research largely overlooks how these models interact with developers in practice and neglects to address when a developer should receive completion suggestions. To tackle this issue, we developed a machine learning model that can accurately predict when to invoke a code completion tool given the code context and available telemetry data. To do so, we collect a dataset of 200k developer interactions with our cross-IDE code completion plugin and train several invocation filtering models. Our results indicate that our small-scale…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
- 🤗AISE-TUDelft/CodeBERTa-ft-coco-1e-05lrmodel· 15 dl15 dl
- 🤗AISE-TUDelft/CodeBERTa-ft-coco-2e-05lrmodel· 1 dl1 dl
- 🤗AISE-TUDelft/CodeBERTa-ft-coco-5e-05lrmodel· 2 dl2 dl
- 🤗AISE-TUDelft/JonBERTa-head-ft-cocomodel· 12 dl12 dl
- 🤗AISE-TUDelft/JonBERTa-head-ft-coco-reinitmodel· 1 dl1 dl
- 🤗AISE-TUDelft/JonBERTa-head-ft-coco-projmodel
- 🤗AISE-TUDelft/JonBERTa-head-ft-coco-proj-reinitmodel
- 🤗AISE-TUDelft/JonBERTa-head-ft-coco-densemodel
- 🤗AISE-TUDelft/JonBERTa-head-ft-coco-dense-reinitmodel· 1 dl1 dl
- 🤗AISE-TUDelft/JonBERTa-head-ft-coco-dense-projmodel· 1 dl1 dl
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
