Where release 1.0 was tuned for use on the common swift (an ideal light logger species) this release sees improvements in which cover expanding the methodology to species which have less than optimal (noisy) light logger data. It provides new methods to correctly estimate parameter ranges (attenuation factors due to preferred secluded habitats).
- introduction of a filter function
stk_filter(), but helpful in visualizing data
- introduction of the
stk_calibrate() function to estimate scale factors for optimization
- correction to the subsetting of the data (see
stk_filter()) to accommodate noisy data
- trap rendering error on intermediate plots, which corrupts optimization
- support for a threshold based (twilight) method when only providing a single
range value
- removed seed from
stk_cluster()
- move to wrappers
- added references to theory papers
- support batch reading of files
- support batch screening of data using
stk_screen_twl()
- verbose feedback on data reading
- allow plotting of data with uneven data steps (in time)
- assign a placeholder “logger” name in the mapping function
stk_map()
- vignette on parallel processing using {multidplyr}
- vignette on optimization improvement strategies
- adding reference into package DESCRIPTION file
- CRAN compliant release
- draft release and working proof-of-concept