A main problem with reproducing machine learning publications is the variance of metric implementations across papers. A lack of standardization leads to different behavior in mechanisms such as checkpointing, learning rate schedulers or early stopping, that will influence the reported results. For example, a complex metric such as Fréchet inception distance (FID) for synthetic image quality evaluation (Heusel et al., 2017) will differ based on the specific interpolation method used.
TorchMetrics - Measuring Reproducibility in PyTorch
N. Detlefsen,Jiri Borovec,Justus Schock,A. Jha,Teddy Koker,Luca Di Liello,Daniel Stancl,Changsheng Quan,Maxim Grechkin,William Falcon
Published 2022 in Journal of Open Source Software
ABSTRACT
PUBLICATION RECORD
- Publication year
2022
- Venue
Journal of Open Source Software
- Publication date
2022-02-11
- Fields of study
Computer Science
- Identifiers
- External record
- Source metadata
Semantic Scholar
CITATION MAP
EXTRACTION MAP
CLAIMS
- No claims are published for this paper.
CONCEPTS
- No concepts are published for this paper.
REFERENCES
Showing 1-7 of 7 references · Page 1 of 1