Writing

[jkarolczak]

• [ ] Dataset
  • [ ] Origin
  • [ ] Train/test/holdout split
  • [ ] Description of statistics
• [ ] Imbalanced data
  • [ ] Distribution of classes
  • [x] Random over- and under- samplings
• [x] Preprocessing - @wtaisner, @annprzy
  • [x] shell
  • [x] uniform
  • [x] kmeans
  • [x] PCA
  • [x] normals
• [ ] Models - @konradszewczyk
  • [ ] TransLoc3D
  • [ ] MinkLoc3DV2
  • [ ] RiConv (?)
  • [ ] Models implemented in MinkowskiEngine
• [ ] Runtime environment - @jkarolczak
  • [ ] MLOps
  • [ ] hardware
  • [ ] MinkowskiEngine
  • [ ] DevOps
  • [ ] Testing, CI
  • [ ] Storage - Azure

[wtaisner]

I would also add, just to remember:

• distribution of classes
• sampling methods

[jkarolczak]

@wtaisner added!

[wtaisner]

I think it's done