Can't use `pam_classify()` and `trainset_write()` without acceleration data

[Rafnuss]

If no acceleration data is available for your track, some functions won't work. With the error message:

Error in pam_classify(pam_data) :
  "date" %in% names(pam$acceleration) is not TRUE

This is the case for

• https://github.com/Rafnuss/Val-Piora-Wheatear
• https://github.com/grhyne/SWWA_Pressure

[Rafnuss]

One way of solving this is to create fake acceleration data and bypass pam_classify() with,

pam_data$acceleration = pam_data$pressure
pam_data$acceleration$obs = 0
pam_data$acceleration$ismig = FALSE

• This would create an acceleration data at the same resolution than the pressure which seems a good choice by default, but you could also have it on anther resolution.
• You will have to label manually all migration activity data.
• Another option would be to create a new function to classify activity from the change of pressure (e.g., changepoint analysis). (https://kiranlda.github.io/PAMLrManual/method.html)