Import2vec: learning embeddings for software libraries
We consider the problem of developing suitable learning representations (embeddings) for library packages that capture semantic similarity among libraries. Such representations are known to improve the performance of downstream learning tasks (e.g. classification) or applications such as contextual search and analogical reasoning.
For a live demo of a contextual search engine built on import2vec, please visit https://www.code-compass.com.
Sun 26 MayDisplayed time zone: Eastern Time (US & Canada) change
11:00 - 11:45
|SCOR: Source Code Retrieval With Semantics and Order|
MSR 2019 Technical PapersPre-print Media Attached
|PathMiner : A Library for Mining of Path-Based Representations of Code|
MSR 2019 Technical Papers
Vladimir Kovalenko TU Delft, Egor Bogomolov Higher School of Economics, JetBrains Research, Timofey Bryksin , Alberto Bacchelli University of ZurichDOI Pre-print Media Attached
|Import2vec: learning embeddings for software libraries|
MSR 2019 Technical PapersPre-print
|Semantic Source Code Models Using Identifier Embeddings|
MSR 2019 Data Showcase
Vasiliki Efstathiou Athens University of Economics and Business, Diomidis Spinellis Athens University of Economics and BusinessPre-print