Skip to content
This repository has been archived by the owner on Mar 17, 2021. It is now read-only.

NiftyNet dev meeting 1st February 2019

csudre edited this page Feb 3, 2019 · 4 revisions

NiftyNet meeting - 1st February 2019

Attending: Tom Ve, Zach, Fernando, Wenqi, Ben, Felix, Carole

Next version ready for release - to happen early March

  • Documentation
  • Model zoo
  • Multiple optimisers in application
  • Random scaling
  • Anti alias in random scaling - same to be put in resize sampler
  • Difficult to keep compatibility 1.12 Cu 7.1

Next stage for release in May

  • Batch norm pull request - work on compatibility with Model Zoo
  • Pull request for CSV reader
  • Multi stage training / Early stopping / Performance handler
  • Need for tests using very small networks

Release process

  • Dzhosh to help BM interested in following process
  • Work plan on the wiki
    • From one release to the next

Additional features

  • Padder on uniform BM
  • Exponential decay rate BM
  • Preprocessing for regression to segmentation Histogram. Quantisation. FB

To keep in mind / Longer term

  • Resampling on GPU
  • Cycle GAN
  • More composibility BM - Isolating data augmentation
  • Application API - Integration into other application
  • 142 Yaml config file to be rekindled with BM
  • Template for issue

Google Summer Code

  • 2 months
  • Provide project by 6th February
Clone this wiki locally