LiJunnan1992 / DivideMix

Code for paper: DivideMix: Learning with Noisy Labels as Semi-supervised Learning
MIT License
529 stars 83 forks source link

why change dimension #52

Open IwGrin opened 1 year ago

IwGrin commented 1 year ago

why cahnge (samples, channels, height, width) to (samples, height, width, channels) in cifar-10 and cifar-100? (transpose((0, 2, 3, 1))) I wonder the last dimension is (samples, channels, height, width) or (samples, height, width, channels) ?

question from a rooooookie

IwGrin commented 1 year ago

to match what