crosszamirski/WS-DINO

WS-DINO: a novel framework to use weak label information in a self-supervised setting to learn phenotypic representations from high-content fluorescent images of cells.

Python

Issues

Unstable training loss
#6 opened a year ago by LiuBodan
0
Unsatisfactory performances from compounds as weakly-supervised label
#4 opened 2 years ago by Willaem
2
CSV files required by illumination correction pipelines
#2 opened 2 years ago by hellowangqian
16
utils module?
#5 opened 2 years ago by ca-ug
1
Weak label selection as Training parameter
#3 opened 2 years ago by Willaem
0
WeightRandomSampler does not work properly while DDP
#1 opened 2 years ago by gabrielpan147
4