DeepRace applies a CNN with class activation mapping to detect data race bugs at file level (81-86% accuracy) and localize buggy lines (66% IoU) on OpenMP and POSIX datasets of over 13,000 files.
Available: ht tps://github.com/eliben/pycparser
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DeepRace: Finding Data Race Bugs via Deep Learning
DeepRace applies a CNN with class activation mapping to detect data race bugs at file level (81-86% accuracy) and localize buggy lines (66% IoU) on OpenMP and POSIX datasets of over 13,000 files.