Denchmark: A bug benchmark of deep learning-related software

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

15 Scopus citations

Abstract

A growing interest in deep learning (DL) has instigated a concomitant rise in DL-related software (DLSW). Therefore, the importance of DLSW quality has emerged as a vital issue. Simultaneously, researchers have found DLSW more complicated than traditional SW and more difficult to debug owing to the black-box nature of DL. These studies indicate the necessity of automatic debugging techniques for DLSW. Although several validated debugging techniques exist for general SW, no such techniques exist for DLSW. There is no standard bug benchmark to validate these automatic debugging techniques. In this study, we introduce a novel bug benchmark for DLSW, Denchmark, consisting of 4, 577 bug reports from 193 popular DLSW projects, collected through a systematic dataset construction process. These DLSW projects are further classified into eight categories: framework, platform, engine, compiler, tool, library, DL-based application, and others. All bug reports in Denchmark contain rich textual information and links with bug-fixing commits, as well as three levels of buggy entities, such as files, methods, and lines. Our dataset aims to provide an invaluable starting point for the automatic debugging techniques of DLSW.

Original languageEnglish
Title of host publicationProceedings - 2021 IEEE/ACM 18th International Conference on Mining Software Repositories, MSR 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages540-544
Number of pages5
ISBN (Electronic)9781728187105
DOIs
StatePublished - May 2021
Event18th IEEE/ACM International Conference on Mining Software Repositories, MSR 2021 - Virtual, Online
Duration: 17 May 202119 May 2021

Publication series

NameProceedings - 2021 IEEE/ACM 18th International Conference on Mining Software Repositories, MSR 2021

Conference

Conference18th IEEE/ACM International Conference on Mining Software Repositories, MSR 2021
CityVirtual, Online
Period17/05/2119/05/21

Keywords

  • Automatic debugging
  • Bug Bench-mark
  • Bug report
  • Deep learning-related software

Fingerprint

Dive into the research topics of 'Denchmark: A bug benchmark of deep learning-related software'. Together they form a unique fingerprint.

Cite this