getProbability: Classify the typical days by searching for external factors...
In rte-antares-rpackage/flowBasedClustering: Cluster 'Flow-Based' Typical Days

Classify the typical days by searching for external factors (e.g. climate factors) explaning the occurence of a particular typical day

1 2	getProbability(climate, cluster, levelsProba = c(1/3, 2/3), extrapolationNA = TRUE)

`climate`	`data.table`, the first column contains the dates (header : Date, format : YYYY-MM-DD), other columns are external variable whose name must be given in the header. Data should be given with a daily time step.
`cluster`	`data.table` output of clusteringTypicalDays or clusterTypicalDaysForOneClass
`levelsProba`	`numeric or list`. if `numeric`, contains a vector of quantiles which will be used for all external variables and classes. if `list`, quantiles are described independenly for each class and each external variable (see examples below)
`extrapolationNA`	`booelan` should NA values of the matrix be extrapolated (`TRUE`) or not (`FALSE`)

# load climate daily time serires
library(data.table)
climate <- fread(system.file("dataset/climate_example.txt",package = "flowBasedClustering"))

# load clustering results (or build them with clusteringTypicalDays function())
clusterTD <- readRDS(system.file("dataset/cluster_example.RDS",package = "flowBasedClustering"))

# get Probability matrices
MatProb <- getProbability(climate, clusterTD)

# run with specified probability levels, and witout auto-filling the NA cases
levelsProba <- list(summerWd = list(FR_load = c(0.5), DE_wind = c(1/3, 2/3), DE_solar = .5),
                    summerWe = list(FR_load = c(0.5, 0.7), DE_wind = c(.5))
)
MatProb <- getProbability(climate, clusterTD, levelsProba = levelsProba, extrapolationNA = FALSE)
MatProb2 <- getProbability(climate, clusterTD, levelsProba = levelsProba, extrapolationNA = TRUE)