Bug #3364
openComplete PAR data acquisition workflow
0%
Description
Reorganize par-data-clean step, replace with generic actors:
- Elementwise: min, max, is.na
- UnivariateDistribution: min, max, range, interquartile range, mean, variance, median, stddev, historgram, densityplot, autocorrelation, partial autocorrelation; bivariate: correlation
- statistical model fit: lm, glm, gam; filter actors
Updated by Derik Barseghian over 16 years ago
I've completed a rough draft of this workflow and emailed a request for comments.
It is very fine grained, and uses generic R actors that I've created, including: univariate threshold test, bivariate threshold test, aggregate, set name, merge, R sqrt, R division.
Or'ing two arrays together, and separating dataframes into components were also common tasks and non-generic R actors were used.
Updated by Derik Barseghian about 16 years ago
After some feedback from Matt, much of the plumbing that made this workflow so large has been removed.
Actors have been made more generic and parameter-driven, and workflow simplified, with an eye towards making it more ecologist-friendly, and actors more reusable.