Skip to content
This repository was archived by the owner on Mar 3, 2020. It is now read-only.

Open Items

David Hill edited this page Oct 4, 2016 · 4 revisions

Open Items

  1. Data structure for change result (rmse + magnitude per band)

  2. Preprocess inputs [statistics, filtering, fmask/tmask]

  3. Handle variable selection (4,6 or 8 coefficients)

  4. Recalculate RMSE after each fit

  5. Verify that we are not inadvertently moving to native python data structures (like lists, tuples, etc) from numpy datatypes. Performance killer

Clone this wiki locally