compareCLASS | R Documentation |
this function takes multiple classifications of the same data and compares the agreement between each, on a timepoint level.
compareCLASS(date, classifications)
date |
datetime in POSIXCT format. See hoopoe$activity$date for example |
classifications |
a dataframe containing the results from different classifications in each column. The classifications must correspond to the same datetimes. |
a dataframe containing date
the raw classifications
each state with the number of times it was used in a classification
whether or not all classes provided the same state
## Import data #data(hoopoe) #start = as.POSIXct("2016-07-01","%Y-%m-%d", tz="UTC") #end = as.POSIXct("2017-06-01","%Y-%m-%d", tz="UTC") #PAM_data= cutPAM(hoopoe,start,end) ## perform one classification using classifyFLAP #classification = classifyFLAP(dta = PAM_data$acceleration, period = 12) ## Put the classification in the same resolution as pressure #class1 = classification2PAM(from = classification$timetable$start, # to = classification$timetable$end, # # because the timetable only contains migration periods # classification = rep_len(1,length(classification$timetable$end)), # addTO = PAM_data$pressure) ## Convert to categories #class1 = ifelse(class1 == classification$migration, "Migration", "Other") ## Perform another classification using pressure difference #class2 = c(0,ifelse(abs(diff(PAM_data$pressure$obs))>2, "Migration", "Other")) ## both classes have been converted to the same time intervals as pressure, #date = PAM_data$pressure$date ## Combine the classifications into a dataframe #classifications = data.frame(flap = class1, # flapping classification # Pdiff = class2) # pressure difference classification #class_comparison = compareCLASS(date=date, # classifications=classifications) #plot(PAM_data$pressure$date, # PAM_data$pressure$obs, # type = "l", # xlab = "Date", # ylab = "Pressure (hPa)", # col = "royalblue3") #points(PAM_data$pressure$date, # PAM_data$pressure$obs, # cex = class_comparison$Migration / 2, # col ="orange", # pch = 16)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.