Skip to main content
. 2022 Feb 18;22(4):1596. doi: 10.3390/s22041596
Algorithm 1: Teacher-student training overview
  • 1:

    Train model on supervised set in burn-in step for 10,000 iterations

  • 2:

    After burn-in duplicate model into teacher and student

  • 3:

    for each training iteration on a set of unsupervised images do

  • 4:

           Teacher generates pseudo-labels on images with weak augmentation

  • 5:

           Student uses pseudo-labels to update network with strong augmentation

  • 6:

           Teacher network refined using EMA from update student network

  • 7:

    end for