License: Creative Commons Attribution 3.0 Unported license (CC BY 3.0)
When quoting this document, please refer to the following
DOI: 10.4230/LIPIcs.ECOOP.2020.15
URN: urn:nbn:de:0030-drops-131726
URL: http://dagstuhl.sunsite.rwth-aachen.de/volltexte/2020/13172/
Go to the corresponding LIPIcs Volume Portal

Lagouvardos, Sifis ; Dolby, Julian ; Grech, Neville ; Antoniadis, Anastasios ; Smaragdakis, Yannis

Static Analysis of Shape in TensorFlow Programs

pdf-format:
LIPIcs-ECOOP-2020-15.pdf (0.6 MB)

Abstract

Machine learning has been widely adopted in diverse science and engineering domains, aided by reusable libraries and quick development patterns. The TensorFlow library is probably the best-known representative of this trend and most users employ the Python API to its powerful back-end. TensorFlow programs are susceptible to several systematic errors, especially in the dynamic typing setting of Python. We present Pythia, a static analysis that tracks the shapes of tensors across Python library calls and warns of several possible mismatches. The key technical aspects are a close modeling of library semantics with respect to tensor shape, and an identification of violations and error-prone patterns. Pythia is powerful enough to statically detect (with 84.62% precision) 11 of the 14 shape-related TensorFlow bugs in the recent Zhang et al. empirical study - an independent slice of real-world bugs.

BibTeX - Entry

@InProceedings{lagouvardos_et_al:LIPIcs:2020:13172,
  author =	{Sifis Lagouvardos and Julian Dolby and Neville Grech and Anastasios Antoniadis and Yannis Smaragdakis},
  title =	{{Static Analysis of Shape in TensorFlow Programs}},
  booktitle =	{34th European Conference on Object-Oriented Programming (ECOOP 2020)},
  pages =	{15:1--15:29},
  series =	{Leibniz International Proceedings in Informatics (LIPIcs)},
  ISBN =	{978-3-95977-154-2},
  ISSN =	{1868-8969},
  year =	{2020},
  volume =	{166},
  editor =	{Robert Hirschfeld and Tobias Pape},
  publisher =	{Schloss Dagstuhl--Leibniz-Zentrum f{\"u}r Informatik},
  address =	{Dagstuhl, Germany},
  URL =		{https://drops.dagstuhl.de/opus/volltexte/2020/13172},
  URN =		{urn:nbn:de:0030-drops-131726},
  doi =		{10.4230/LIPIcs.ECOOP.2020.15},
  annote =	{Keywords: Python, TensorFlow, static analysis, Doop, Wala}
}

Keywords: Python, TensorFlow, static analysis, Doop, Wala
Collection: 34th European Conference on Object-Oriented Programming (ECOOP 2020)
Issue Date: 2020
Date of publication: 06.11.2020
Supplementary Material: ECOOP 2020 Artifact Evaluation approved artifact available at https://doi.org/10.4230/DARTS.6.2.6.

DROPS-Home | Fulltext Search | Imprint | Privacy Published by LZI