[Re] Reproducibility study of "Optimizing Deep Learning Inference on Embedded Systems Through Adaptive Model Selection"

[DrejcPesjak]

Original article: Optimizing Deep Learning Inference on Embedded Systems Through Adaptive Model Selection

PDF URL: https://github.com/HitkoDev/energy-efficient-ml/blob/master/article.pdf
Metadata URL: https://github.com/HitkoDev/energy-efficient-ml/blob/master/report/metadata.yaml
Code URL: https://github.com/HitkoDev/energy-efficient-ml/tree/master/code

Scientific domain: Machine Learning
Programming language: Python
Suggested editor:

[rougier]

Thanks for your submission and sorry for the delay. Having read rapidly your article, do you conclude the original paper is not replicable (which it's fine to publish for us). I'm asking mostly because we'll need extra precaution during the review and possibly we'll ask original authors if they can help.

[DrejcPesjak]

We do in fact conclude the original article is not reproducible with information presented therein.

[rougier]

Sorry (again) for long delay, for some reason your submission disapear from my notifications.
Are your still ok we proceed with the review?