Supplementary for "How Composite Metamorphic Relations Enhance Test Effectiveness of DNN Testing: An Empirical Study"
Wu, Huayao; Wang, Peng; Hu, Shengyou; Niu, Xintao; Nie, Changhai; Chen, Tsung Yueh;
Supplementary for "How Composite Metamorphic Relations Enhance Test Effectiveness of DNN Testing: An Empirical Study"
This repository is supplementary to the paper "How Composite Metamorphic Relations Enhance Test Effectiveness of DNN Testing: An Empirical Study" containing: a data/ folder (source test images, CMR-generated follow-up images — full follow-up set is very large ~78 TB so only sampled human_validation images are provided), a models/ folder (pretrained DNN weights and codes to be placed there), and a results/ folder (model predictions, SelfOracle validity outputs, failures/faults, extracted feature tensors for ΔM analysis, and sampled indices for data‑augmentation evaluation).
- Nanjing University China (People's Republic of)
- Swinburne University of Technology Australia
selected citations These citations are derived from selected sources.This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).0 popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.Average influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).Average impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.Average
selected citations
Citations provided by BIP!
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
Citations provided by BIP!popularityPopularity provided by BIP!
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
Popularity provided by BIP! 0
Average
Average
Average
Green