Skip to content

Latest commit

 

History

History
39 lines (34 loc) · 2.01 KB

File metadata and controls

39 lines (34 loc) · 2.01 KB

TODO

  • Plotting script to combine NSD, velocity, niche size and dissim in plots for every individual
  • Need to come up with "final" and well-justified set of niche variables
    • and be able to annotate with those vars...
      *May require modes to join_annos.r script
  • Think about some sort of spatial size measure on same weekly time scales - do niche dynamics conform to spatial dynamics?
    *Find a way to aggregate/summarize relationships across individuals so I can make inter-species comparisons
  • Put conda info in the README
    • export relevant .yml files
    • include instructions for creating conda env from yml

OLD TODO

  • Finish annotate_tracks.r script *Confirm ability to pass variables from ctfs to annotator. *rstoat throwing error with particular products
  • Next script is to feed annotated data into MVNH niche estimator

Activity log

Date Activity
2021-07-20 Pilot WF completed up to annotator call
2021-08-06 Setting aside old wf dev b/c we have a fully-annotated crane dataset, proceeding with new wf building from that
2021-08-16 Crane workflow built through flexible joining of annotations and time-dynamic niche size estimation.
2021-08-18 Working scripts to produce NSD, velocity, ncihe size, niche dissim!
2021-09/10-all Missing logs
2021-10-25 Trying to incorporate niche mixtures into the workflow, started building a script to segment for all seasons from ctf
2021-20-27 Script that gets ind and season-specific mean and var from full data set - returns csv per sp

Notes

  • Cranes pre-anotated by STOAT team at Map of Life.
  • Questions:
    • What's the baseline for dissimilarity calculations?
    • Common set of niche variables?
  • For ind- season-specific mean and var script, note that A. paradesius apparently correctly returns df/cdv of nrow=0

OLD NOTES

  • The anotator relies on breaking data up into chunks of 1000 - this could probably be parallelized easily for large datasets
    • Consider writing a version of the annotate_tracks.r script for the HPC