dtwSat (version 0.2.8)

twdtwCrossValidate: Cross Validate temporal patterns

Description

Splits the set of time series into training and validation and computes accuracy metrics. The function uses stratified sampling and a simple random sampling for each stratum. For each data partition this function performs a TWDTW analysis and returns the Overall Accuracy, User's Accuracy, Produce's Accuracy, error matrix (confusion matrix), and a data.frame with the classification (Predicted), the reference classes (Reference), and the results of the TWDTW analysis.

Usage

# S4 method for twdtwTimeSeries
twdtwCrossValidate(object, times, p, ...)

Arguments

object

An object of class twdtwTimeSeries.

times

Number of partitions to create.

p

The percentage of data that goes to training. See createDataPartition for details.

...

Other arguments to be passed to createPatterns and to twdtwApply.

Author

Victor Maus,

References

Maus:2019dtwSat

Maus:2016dtwSat

Examples

if (FALSE) {
# Data folder 
data_folder = system.file("lucc_MT/data", package = "dtwSat")

# Read dates 
dates = scan(paste(data_folder,"timeline", sep = "/"), what = "dates")

# Read raster time series 
evi = brick(paste(data_folder,"evi.tif", sep = "/"))
raster_timeseries = twdtwRaster(evi, timeline = dates)

# Read field samples 
field_samples = read.csv(paste(data_folder,"samples.csv", sep = "/")) 
table(field_samples[["label"]])

# Read field samples projection 
proj_str = scan(paste(data_folder,"samples_projection", sep = "/"), 
     what = "character")

# Get sample time series from raster time series 
field_samples_ts = getTimeSeries(raster_timeseries, 
     y = field_samples, proj4string = proj_str)
field_samples_ts

# Run cross validation
set.seed(1)
# Define TWDTW weight function 
log_fun = logisticWeight(alpha=-0.1, beta=50) 
cross_validation = twdtwCrossValidate(field_samples_ts, times=3, p=0.1, 
                          freq = 8, formula = y ~ s(x, bs="cc"), weight.fun = log_fun)
cross_validation

summary(cross_validation)

plot(cross_validation)

twdtwXtable(cross_validation)

twdtwXtable(cross_validation, show.overall=FALSE)

}