The authors reproduce the TRUNK neural network across three datasets, find that missing training details cause large accuracy gaps, and extend existing reproducibility guidelines with sensitivity analysis and minimal dependency manifests.
Challenges and practices of deep learning model reengi- neering: A case study on computer vision,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Improving the Reproducibility of Deep Learning Software: An Initial Investigation through a Case Study Analysis
The authors reproduce the TRUNK neural network across three datasets, find that missing training details cause large accuracy gaps, and extend existing reproducibility guidelines with sensitivity analysis and minimal dependency manifests.