search

lucidrains / med-seg-diff-pytorch

Implementation of MedSegDiff in Pytorch - SOTA medical segmentation using DDPM and filtering of features in fourier space
MIT License
209 stars 28 forks source link

fixes #7

Closed bf2harven closed 1 year ago

bf2harven commented 1 year ago
  1. added generic dataset loader
  2. fix driver.py timesteps=args.timesteps instead of args.epochs
  3. add save model in driver.py
  4. add load model in driver.py
  5. fix wandb image logger that requires channels last or no channels.
  6. added ability to evaluate and save model every n epochs to train faster.
lucidrains commented 1 year ago

@bf2harven thank you!