tests/testthat/test_santaR_fit.R
In adwolfer/santaR: Short Asynchronous Time-Series Analysis
context('santaR_fit()')
## Input data
# G1 ind_1 ind_2 ind_3
# G2 ind_4 ind_5 ind_6
# G3 ind_7 only NA (trigger no individual left (logical))
# G4 ind_8 3 tp < 4 (trigger no individual left (logical as empty matrix))
# G5 ind_9 4 tp < df (trigger no individual left (logical as empty matrix))
input_matrix <- data.frame(matrix(nrow=9, ncol=7), stringsAsFactors = FALSE)
colnames(input_matrix) <- c(0, 4, 8, 12, 24, 48, 72)
rownames(input_matrix) <- c('ind_1','ind_2','ind_3','ind_4','ind_5','ind_6','ind_7','ind_8','ind_9')
input_matrix['ind_1',] <- c(-0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763)
input_matrix['ind_2',] <- c(-0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509, as.numeric(NA))
input_matrix['ind_3',] <- c(-0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981)
input_matrix['ind_4',] <- c(-0.4101337, -0.3001967, -0.36492017, -0.26852436, -0.3401291, -0.3251096, -0.2837459)
input_matrix['ind_5',] <- c(-0.4101337, -0.3399216, -0.41013374, -0.32758869, as.numeric(NA), -0.3632320, as.numeric(NA))
input_matrix['ind_6',] <- c(-0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968)
input_matrix['ind_7',] <- c(as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA))
input_matrix['ind_8',] <- c(as.numeric(NA), -0.2975866, as.numeric(NA), -0.18805090, -0.2293608, as.numeric(NA), as.numeric(NA))
input_matrix['ind_9',] <- c(as.numeric(NA), -0.2975866, as.numeric(NA), -0.18805090, -0.2293608, -0.3167715, as.numeric(NA))
input_grouping <- data.frame(matrix(data=c('ind_1','G1','ind_2','G1','ind_3','G1','ind_4','G2','ind_5','G2','ind_6','G2','ind_7','G3','ind_8','G4','ind_9','G5'), nrow=9, ncol=2, byrow=TRUE, dimnames = list(c(),c('ind','group'))), stringsAsFactors=FALSE)
# both ind are (#tp < df), trigger total mean curve (#tp < df)
input_matrix2 <- data.frame(matrix(nrow=2, ncol=7), stringsAsFactors = FALSE)
colnames(input_matrix2) <- c(0, 4, 8, 12, 24, 48, 72)
rownames(input_matrix2) <- c('ind_1','ind_2')
input_matrix2['ind_1',] <- c(-0.4101337, as.numeric(NA), 0.65175696, as.numeric(NA), as.numeric(NA), -0.2975552, -0.3135763)
input_matrix2['ind_2',] <- c(as.numeric(NA), -0.2975866, as.numeric(NA), -0.18805090, -0.2293608, -0.3167715, as.numeric(NA))
# no individual in the group 'dim(groupData.in)[1]==0' shouldn't be possible to trigger as matrix is only updated to data.frame if an individual is added
# no groupMean curve (group data tp < df ) shouldn't be possible to trigger as it's only checked if groupData.in is a data.frame (so minimum one ind)
## Expected
# with grouping
expected_obj <- list()
expected_obj$properties <- list()
expected_obj$properties$df <- 5
expected_obj$properties$CBand$status <- FALSE
expected_obj$properties$CBand$nBoot <- NA
expected_obj$properties$CBand$alpha <- NA
expected_obj$properties$pval.dist$status <- FALSE
expected_obj$properties$pval.dist$nPerm <- NA
expected_obj$properties$pval.dist$alpha <- NA
expected_obj$properties$pval.fit$status <- FALSE
expected_obj$properties$pval.fit$nPerm <- NA
expected_obj$properties$pval.fit$alpha <- NA
expected_obj$groups <- list()
expected_obj$groups$G1$rejectedInd <- list()
expected_obj$groups$G1$curveInd <- list()
expected_obj$groups$G1$curveInd$ind_1 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763), df=5)
expected_obj$groups$G1$curveInd$ind_2 <- stats::smooth.spline(c(0,4,8,12,24,48), c(-0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509), df=5)
expected_obj$groups$G1$curveInd$ind_3 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981), df=5)
expected_obj$groups$G1$groupMeanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.414391776666666656, 1.351222410000000096, 0.755240270000000047, 0.078953003333333313, -0.359661066666666696, -0.324312383333333343, -0.455263819999999986), df=7)
expected_obj$groups$G1$point.in <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,0,4,8,12,24,48,72, -0.41013370,3.77734620,0.65175696,-0.04287806,-0.27873830,-0.29755520,-0.31357630,-0.41013370,2.66841520,-0.08955276,-0.18796505,-0.31772180,-0.32895090,-0.41013370,2.49843700,-0.15771390,-0.24400003,-0.33381800,-0.33708020,-0.33309810), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G1$point.pred <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72, 0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G1$groupData.in <- input_matrix[1:3,]
expected_obj$groups$G1$groupData.pred <- data.frame(matrix(data=c(0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814), ncol=7, dimnames=list(c('ind_1','ind_2','ind_3'),c()), byrow=TRUE), stringsAsFactors=FALSE)
colnames(expected_obj$groups$G1$groupData.pred) <- colnames(input_matrix)
expected_obj$groups$G2$rejectedInd <- list()
expected_obj$groups$G2$curveInd <- list()
expected_obj$groups$G2$curveInd$ind_4 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3001967, -0.3649202, -0.2685244, -0.3401291, -0.3251096, -0.2837459), df=5)
expected_obj$groups$G2$curveInd$ind_5 <- stats::smooth.spline(c(0,4,8,12,48), c(-0.4101337, -0.3399216, -0.4101337, -0.3275887, -0.363232), df=5)
expected_obj$groups$G2$curveInd$ind_6 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968), df=5)
expected_obj$groups$G2$groupMeanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.39910886666666667, -0.35468173333333330, -0.36584719999999998, -0.32711873333333336, -0.27520023333333332, -0.36585533333333337, -0.44383700000000004), df=7)
expected_obj$groups$G2$point.in <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,48,0,4,8,12,24,48,72, -0.4101337,-0.3001967,-0.36492017,-0.26852436,-0.3401291,-0.3251096,-0.2837459,-0.4101337,-0.3399216,-0.41013374,-0.32758869,-0.3632320,-0.4101337,-0.3455376,-0.41013374,-0.33124182,-0.3650980,-0.4101337,-0.2877968), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G2$point.pred <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72, -0.3890455,-0.3450127,-0.3189128,-0.2994064,-0.3304448,-0.3262453,-0.283692,-0.4101337,-0.3399216,-0.4101337,-0.3275887,-0.1319989,-0.363232,-0.7591037,-0.3981474,-0.3791109,-0.3684951,-0.3543611,-0.363157,-0.4080887,-0.2887153), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G2$groupData.in <- input_matrix[4:6,]
expected_obj$groups$G2$groupData.pred <- data.frame(matrix(data=c(-0.3890455,-0.3450127,-0.3189128,-0.2994064,-0.3304448,-0.3262453,-0.283692,-0.4101337,-0.3399216,-0.4101337,-0.3275887,-0.1319989,-0.363232,-0.7591037,-0.3981474,-0.3791109,-0.3684951,-0.3543611,-0.363157,-0.4080887,-0.2887153), ncol=7, dimnames=list(c('ind_4','ind_5','ind_6'),c()), byrow=TRUE), stringsAsFactors=FALSE)
colnames(expected_obj$groups$G2$groupData.pred) <- colnames(input_matrix)
expected_obj$groups$G3$rejectedInd <- list()
expected_obj$groups$G3$curveInd <- NA
expected_obj$groups$G3$groupMeanCurve <- NA
expected_obj$groups$G3$point.in <- NA
expected_obj$groups$G3$point.pred <- NA
expected_obj$groups$G3$groupData.in <- NA
expected_obj$groups$G3$groupData.pred <- NA
expected_obj$groups$G4$rejectedInd <- list()
expected_obj$groups$G4$rejectedInd$ind_8 <- input_matrix[8,]
expected_obj$groups$G4$curveInd <- NA
expected_obj$groups$G4$groupMeanCurve <- NA
expected_obj$groups$G4$point.in <- NA
expected_obj$groups$G4$point.pred <- NA
expected_obj$groups$G4$groupData.in <- NA
expected_obj$groups$G4$groupData.pred <- NA
expected_obj$groups$G5$rejectedInd <- list()
expected_obj$groups$G5$rejectedInd$ind_9 <- input_matrix[9,]
expected_obj$groups$G5$curveInd <- NA
expected_obj$groups$G5$groupMeanCurve <- NA
expected_obj$groups$G5$point.in <- NA
expected_obj$groups$G5$point.pred <- NA
expected_obj$groups$G5$groupData.in <- NA
expected_obj$groups$G5$groupData.pred <- NA
expected_obj$general <- list()
expected_obj$general$inputData <- input_matrix[-7,]
expected_obj$general$grouping <- input_grouping
expected_obj$general$cleanData.in <- input_matrix[1:6,]
expected_obj$general$cleanData.pred <- rbind(expected_obj$groups$G1$groupData.pred, expected_obj$groups$G2$groupData.pred)
expected_obj$general$pval.curveCorr <- NA
expected_obj$general$pval.dist <- NA
expected_obj$general$pval.dist.l <- NA
expected_obj$general$pval.dist.u <- NA
expected_obj$general$pval.fit <- NA
expected_obj$general$pval.fit.l <- NA
expected_obj$general$pval.fit.u <- NA
expected_obj$general$meanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.0076414549999999941, 0.4982703383333333402, 0.1946965350000000039, -0.1240828650000000144, -0.3174306499999999809, -0.3450838583333333265, -0.4495504100000000114), df=7)
# without grouping
expected_obj_noGrp <- list()
expected_obj_noGrp$properties <- list()
expected_obj_noGrp$properties$df <- 5
expected_obj_noGrp$properties$CBand$status <- FALSE
expected_obj_noGrp$properties$CBand$nBoot <- NA
expected_obj_noGrp$properties$CBand$alpha <- NA
expected_obj_noGrp$properties$pval.dist$status <- FALSE
expected_obj_noGrp$properties$pval.dist$nPerm <- NA
expected_obj_noGrp$properties$pval.dist$alpha <- NA
expected_obj_noGrp$properties$pval.fit$status <- FALSE
expected_obj_noGrp$properties$pval.fit$nPerm <- NA
expected_obj_noGrp$properties$pval.fit$alpha <- NA
expected_obj_noGrp$groups <- list()
expected_obj_noGrp$groups[[1]] <- list()
expected_obj_noGrp$groups[[1]]$rejectedInd <- list()
expected_obj_noGrp$groups[[1]]$rejectedInd[[7]] <- input_matrix[8,]
names(expected_obj_noGrp$groups[[1]]$rejectedInd)[[7]] <- 'ind_8'
expected_obj_noGrp$groups[[1]]$rejectedInd[[8]] <- input_matrix[9,]
names(expected_obj_noGrp$groups[[1]]$rejectedInd)[[8]] <- 'ind_9'
expected_obj_noGrp$groups[[1]]$curveInd <- list()
expected_obj_noGrp$groups[[1]]$curveInd$ind_1 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_2 <- stats::smooth.spline(c(0,4,8,12,24,48), c(-0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_3 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_4 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3001967, -0.3649202, -0.2685244, -0.3401291, -0.3251096, -0.2837459), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_5 <- stats::smooth.spline(c(0,4,8,12,48), c(-0.4101337, -0.3399216, -0.4101337, -0.3275887, -0.363232), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_6 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968), df=5)
expected_obj_noGrp$groups[[1]]$groupMeanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.0076414616666666572, 0.4982703316666666771, 0.1946965333333333381, -0.1240828566666666716, -0.3174306383333333481, -0.3450838616666666581, -0.4495504083333333178), df=7)
expected_obj_noGrp$groups[[1]]$point.in <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,48,0,4,8,12,24,48,72, -0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763, -0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509, -0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981, -0.4101337, -0.3001967, -0.3649202, -0.2685244, -0.3401291, -0.3251096, -0.2837459, -0.4101337, -0.3399216, -0.4101337, -0.3275887, -0.363232, -0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj_noGrp$groups[[1]]$point.pred <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72, 0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814,-0.38904549,-0.34501275,-0.3189128,-0.29940637,-0.33044479,-0.32624534,-0.28369199,-0.4101337,-0.3399216,-0.41013374,-0.32758869,-0.13199885,-0.363232,-0.75910373,-0.39814737,-0.37911089,-0.36849507,-0.35436109,-0.36315699,-0.40808868,-0.28871527), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj_noGrp$groups[[1]]$groupData.in <- input_matrix[1:6,]
expected_obj_noGrp$groups[[1]]$groupData.pred <- data.frame(matrix(data=c(0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814,-0.38904549,-0.34501275,-0.3189128,-0.29940637,-0.33044479,-0.32624534,-0.28369199,-0.4101337,-0.3399216,-0.41013374,-0.32758869,-0.13199885,-0.363232,-0.75910373,-0.39814737,-0.37911089,-0.36849507,-0.35436109,-0.36315699,-0.40808868,-0.28871527), ncol=7, dimnames=list(c('ind_1','ind_2','ind_3','ind_4','ind_5','ind_6'),c()), byrow=TRUE), stringsAsFactors=FALSE)
colnames(expected_obj_noGrp$groups[[1]]$groupData.pred) <- colnames(input_matrix)
expected_obj_noGrp$general <- list()
expected_obj_noGrp$general$inputData <- input_matrix[-7,]
expected_obj_noGrp$general$grouping <- NA
expected_obj_noGrp$general$cleanData.in <- input_matrix[1:6,]
expected_obj_noGrp$general$cleanData.pred <- expected_obj_noGrp$groups[[1]]$groupData.pred
expected_obj_noGrp$general$pval.curveCorr <- NA
expected_obj_noGrp$general$pval.dist <- NA
expected_obj_noGrp$general$pval.dist.l <- NA
expected_obj_noGrp$general$pval.dist.u <- NA
expected_obj_noGrp$general$pval.fit <- NA
expected_obj_noGrp$general$pval.fit.l <- NA
expected_obj_noGrp$general$pval.fit.u <- NA
expected_obj_noGrp$general$meanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.0076414616666666572, 0.4982703316666666771, 0.1946965333333333381, -0.1240828566666666716, -0.3174306383333333481, -0.3450838616666666581, -0.4495504083333333178), df=7)
# no total mean curve calculation
expected_obj_2 <- list()
expected_obj_2$properties <- list()
expected_obj_2$properties$df <- 5
expected_obj_2$properties$CBand$status <- FALSE
expected_obj_2$properties$CBand$nBoot <- NA
expected_obj_2$properties$CBand$alpha <- NA
expected_obj_2$properties$pval.dist$status <- FALSE
expected_obj_2$properties$pval.dist$nPerm <- NA
expected_obj_2$properties$pval.dist$alpha <- NA
expected_obj_2$properties$pval.fit$status <- FALSE
expected_obj_2$properties$pval.fit$nPerm <- NA
expected_obj_2$properties$pval.fit$alpha <- NA
expected_obj_2$groups <- list()
expected_obj_2$groups[[1]] <- list()
expected_obj_2$groups[[1]]$rejectedInd <- list()
expected_obj_2$groups[[1]]$rejectedInd[[1]] <- input_matrix2[1,]
names(expected_obj_2$groups[[1]]$rejectedInd)[[1]] <- 'ind_1'
expected_obj_2$groups[[1]]$rejectedInd[[2]] <- input_matrix2[2,]
names(expected_obj_2$groups[[1]]$rejectedInd)[[2]] <- 'ind_2'
expected_obj_2$groups[[1]]$curveInd <- NA
expected_obj_2$groups[[1]]$groupMeanCurve <- NA
expected_obj_2$groups[[1]]$point.in <- NA
expected_obj_2$groups[[1]]$point.pred <- NA
expected_obj_2$groups[[1]]$groupData.in <- NA
expected_obj_2$groups[[1]]$groupData.pred <- NA
expected_obj_2$general <- list()
expected_obj_2$general$inputData <- input_matrix2
expected_obj_2$general$grouping <- NA
expected_obj_2$general$cleanData.in <- matrix(ncol=7,nrow=0)
expected_obj_2$general$cleanData.pred <- matrix(ncol=7,nrow=0)
expected_obj_2$general$pval.curveCorr <- NA
expected_obj_2$general$pval.dist <- NA
expected_obj_2$general$pval.dist.l <- NA
expected_obj_2$general$pval.dist.u <- NA
expected_obj_2$general$pval.fit <- NA
expected_obj_2$general$pval.fit.l <- NA
expected_obj_2$general$pval.fit.u <- NA
expected_obj_2$general$meanCurve <- NA
test_that('default values, with grouping', {
# expected_fit
expected_fit <- expected_obj
# G1 G2 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
exp_curveInd_G1 <- expected_fit$groups$G1$curveInd
exp_curveInd_G2 <- expected_fit$groups$G2$curveInd
exp_grpMnCurve_G1 <- expected_fit$groups$G1$groupMeanCurve
exp_grpMnCurve_G2 <- expected_fit$groups$G2$groupMeanCurve
exp_meanCurve <- expected_fit$general$meanCurve
expected_fit$groups$G1$curveInd <- NULL
expected_fit$groups$G2$curveInd <- NULL
expected_fit$groups$G1$groupMeanCurve <- NULL
expected_fit$groups$G2$groupMeanCurve <- NULL
expected_fit$general$meanCurve <- NULL
# expected message
expected_message <- c("No individuals left in group #3 once data is imported (remove NA rows and columns)\n", "No individuals left in group #3\n", "ind_8 - #tp < 4\n", "No individuals left in group #4\n", "ind_9 - #tp(4) < df(5)\n", "No individuals left in group #5\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=input_grouping, verbose=TRUE))
result_fit_noV <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=input_grouping, verbose=FALSE))
# first check that verbose and no verbose match
expect_equal(result_fit$result, result_fit_noV$result)
# G1 G2 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
entries_to_check <- c("x","y","w","yin","tol","data","no.weights","lev","cv.crit","pen.crit","crit","df","spar","ratio","lambda","iparms","auxM","fit")
tryCatch({
tmp_curveInd_G1 <- result_fit$result$groups$G1$curveInd
tmp_curveInd_G2 <- result_fit$result$groups$G2$curveInd
tmp_grpMnCurve_G1 <- result_fit$result$groups$G1$groupMeanCurve
tmp_grpMnCurve_G2 <- result_fit$result$groups$G2$groupMeanCurve
tmp_meanCurve <- result_fit$result$general$meanCurve
result_fit$result$groups$G1$curveInd <- NULL
result_fit$result$groups$G2$curveInd <- NULL
result_fit$result$groups$G1$groupMeanCurve <- NULL
result_fit$result$groups$G2$groupMeanCurve <- NULL
result_fit$result$general$meanCurve <- NULL
},
error=function(cond) {
## catch, do nothing
tmp_curveInd_G1 <- NA
tmp_curveInd_G2 <- NA
tmp_grpMnCurve_G1 <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_grpMnCurve_G2 <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_meanCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
}
)
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check curveInd (without function call)
expect_equal(length(tmp_curveInd_G1), length(exp_curveInd_G1))
expect_equal(length(tmp_curveInd_G2), length(exp_curveInd_G2))
# curveInd G1
if (length(tmp_curveInd_G1) == length(exp_curveInd_G1)) {
for (i in seq_along(exp_curveInd_G1)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd_G1[[i]][[k]], exp_curveInd_G1[[i]][[k]], tolerance=1e-5)
}
}
}
# curveInd G2
if (length(tmp_curveInd_G2) == length(exp_curveInd_G2)) {
for (i in seq_along(exp_curveInd_G2)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd_G2[[i]][[k]], exp_curveInd_G2[[i]][[k]], tolerance=1e-5)
}
}
}
# Check groupMeanCurve (without function call)
# groupMeanCurve G1
if (!all(is.na(tmp_grpMnCurve_G1))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve_G1[[k]], exp_grpMnCurve_G1[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve_G1, exp_grpMnCurve_G1, tolerance=1e-5)
}
# groupMeanCurve G2
if (!all(is.na(tmp_grpMnCurve_G2))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve_G2[[k]], exp_grpMnCurve_G2[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve_G2, exp_grpMnCurve_G2, tolerance=1e-5)
}
# Check total meanCurve (without function call)
if (!all(is.na(tmp_meanCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_meanCurve[[k]], exp_meanCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_meanCurve, exp_meanCurve, tolerance=1e-5)
}
# Check result messages
expect_equal(length(result_fit$messages), 6)
expect_equal(result_fit$messages, expected_message)
# Check no message without verbose
expect_equal(length(result_fit_noV$messages), 0)
})
test_that('default values, no grouping', {
# expected_fit
expected_fit <- expected_obj_noGrp
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
exp_curveInd <- expected_fit$groups[[1]]$curveInd
exp_grpMnCurve <- expected_fit$groups[[1]]$groupMeanCurve
exp_meanCurve <- expected_fit$general$meanCurve
expected_fit$groups[[1]]$curveInd <- NULL
expected_fit$groups[[1]]$groupMeanCurve <- NULL
expected_fit$general$meanCurve <- NULL
# expected message
expected_message <- c("ind_8 - #tp < 4\n", "ind_9 - #tp(4) < df(5)\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=NA, verbose=TRUE))
result_fit_noV <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=NA, verbose=FALSE))
# first check that verbose and no verbose match
expect_equal(result_fit$result, result_fit_noV$result)
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
entries_to_check <- c("x","y","w","yin","tol","data","no.weights","lev","cv.crit","pen.crit","crit","df","spar","ratio","lambda","iparms","auxM","fit")
tryCatch({
tmp_curveInd <- result_fit$result$groups[[1]]$curveInd
tmp_grpMnCurve <- result_fit$result$groups[[1]]$groupMeanCurve
tmp_meanCurve <- result_fit$result$general$meanCurve
result_fit$result$groups[[1]]$curveInd <- NULL
result_fit$result$groups[[1]]$groupMeanCurve <- NULL
result_fit$result$general$meanCurve <- NULL
},
error=function(cond) {
## catch, do nothing
tmp_curveInd <- NA
tmp_grpMnCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_meanCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
}
)
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check curveInd (without function call)
expect_equal(length(tmp_curveInd), length(exp_curveInd))
# curveInd G1
if (length(tmp_curveInd) == length(exp_curveInd)) {
for (i in seq_along(exp_curveInd)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd[[i]][[k]], exp_curveInd[[i]][[k]], tolerance=1e-5)
}
}
}
# Check groupMeanCurve (without function call)
# groupMeanCurve G1
if (!all(is.na(tmp_grpMnCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve[[k]], exp_grpMnCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve, exp_grpMnCurve, tolerance=1e-5)
}
# Check total meanCurve (without function call)
if (!all(is.na(tmp_meanCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_meanCurve[[k]], exp_meanCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_meanCurve, exp_meanCurve, tolerance=1e-5)
}
# Check result messages
expect_equal(length(result_fit$messages), 2)
expect_equal(result_fit$messages, expected_message)
# Check no message without verbose
expect_equal(length(result_fit_noV$messages), 0)
})
test_that('no total mean curve calculation possible', {
# expected_fit
expected_fit <- expected_obj_2
# expected message
expected_message <- c("ind_1 - #tp(4) < df(5)\n", "ind_2 - #tp(4) < df(5)\n", "No individuals left in group #1\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(input_matrix2, df=5, grouping=NA, verbose=TRUE))
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check result messages
expect_equal(length(result_fit$messages), 3)
expect_equal(result_fit$messages, expected_message)
})
test_that('empty row and col in inputMatrix', {
# empty row (ind_7) & empty col (100) should be removed automatically (use expected_obj_noGrp)
# input
tmp_matrix <- cbind(input_matrix, "100"=c(NA,NA,NA,NA,NA,NA,NA,NA,NA))
# expected_fit
expected_fit <- expected_obj_noGrp
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
exp_curveInd <- expected_fit$groups[[1]]$curveInd
exp_grpMnCurve <- expected_fit$groups[[1]]$groupMeanCurve
exp_meanCurve <- expected_fit$general$meanCurve
expected_fit$groups[[1]]$curveInd <- NULL
expected_fit$groups[[1]]$groupMeanCurve <- NULL
expected_fit$general$meanCurve <- NULL
# expected message
expected_message <- c("ind_8 - #tp < 4\n", "ind_9 - #tp(4) < df(5)\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(tmp_matrix, df=5, grouping=NA, verbose=TRUE))
result_fit_noV <- evaluate_promise(santaR_fit(tmp_matrix, df=5, grouping=NA, verbose=FALSE))
# first check that verbose and no verbose match
expect_equal(result_fit$result, result_fit_noV$result)
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
entries_to_check <- c("x","y","w","yin","tol","data","no.weights","lev","cv.crit","pen.crit","crit","df","spar","ratio","lambda","iparms","auxM","fit")
tryCatch({
tmp_curveInd <- result_fit$result$groups[[1]]$curveInd
tmp_grpMnCurve <- result_fit$result$groups[[1]]$groupMeanCurve
tmp_meanCurve <- result_fit$result$general$meanCurve
result_fit$result$groups[[1]]$curveInd <- NULL
result_fit$result$groups[[1]]$groupMeanCurve <- NULL
result_fit$result$general$meanCurve <- NULL
},
error=function(cond) {
## catch, do nothing
tmp_curveInd <- NA
tmp_grpMnCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_meanCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
}
)
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check curveInd (without function call)
expect_equal(length(tmp_curveInd), length(exp_curveInd))
# curveInd G1
if (length(tmp_curveInd) == length(exp_curveInd)) {
for (i in seq_along(exp_curveInd)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd[[i]][[k]], exp_curveInd[[i]][[k]], tolerance=1e-5)
}
}
}
# Check groupMeanCurve (without function call)
# groupMeanCurve G1
if (!all(is.na(tmp_grpMnCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve[[k]], exp_grpMnCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve, exp_grpMnCurve, tolerance=1e-5)
}
# Check total meanCurve (without function call)
if (!all(is.na(tmp_meanCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_meanCurve[[k]], exp_meanCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_meanCurve, exp_meanCurve, tolerance=1e-5)
}
# Check result messages
expect_equal(length(result_fit$messages), 2)
expect_equal(result_fit$messages, expected_message)
# Check no message without verbose
expect_equal(length(result_fit_noV$messages), 0)
})
test_that('raise error', {
# df not numeric
expect_error(santaR_fit(input_matrix, df='notNumeric', grouping=input_grouping, verbose=FALSE), "Check input, 'df' should be numeric", fixed=TRUE)
# less than 4 tp
expect_error(santaR_fit(input_matrix[1:3], df=3, grouping=NA, verbose=FALSE), "Check input, a minimum of 4 unique time-points are required to fit a smooth.splines", fixed=TRUE)
# tp < df
expect_error(santaR_fit(input_matrix, df=10, grouping=NA, verbose=FALSE), "Check input, the number of unique time-points (7) cannot be less than the requested df (10)", fixed=TRUE)
})
adwolfer/santaR documentation built on March 10, 2024, 5:28 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
context('santaR_fit()')
## Input data
# G1 ind_1 ind_2 ind_3
# G2 ind_4 ind_5 ind_6
# G3 ind_7 only NA (trigger no individual left (logical))
# G4 ind_8 3 tp < 4 (trigger no individual left (logical as empty matrix))
# G5 ind_9 4 tp < df (trigger no individual left (logical as empty matrix))
input_matrix <- data.frame(matrix(nrow=9, ncol=7), stringsAsFactors = FALSE)
colnames(input_matrix) <- c(0, 4, 8, 12, 24, 48, 72)
rownames(input_matrix) <- c('ind_1','ind_2','ind_3','ind_4','ind_5','ind_6','ind_7','ind_8','ind_9')
input_matrix['ind_1',] <- c(-0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763)
input_matrix['ind_2',] <- c(-0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509, as.numeric(NA))
input_matrix['ind_3',] <- c(-0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981)
input_matrix['ind_4',] <- c(-0.4101337, -0.3001967, -0.36492017, -0.26852436, -0.3401291, -0.3251096, -0.2837459)
input_matrix['ind_5',] <- c(-0.4101337, -0.3399216, -0.41013374, -0.32758869, as.numeric(NA), -0.3632320, as.numeric(NA))
input_matrix['ind_6',] <- c(-0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968)
input_matrix['ind_7',] <- c(as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA), as.numeric(NA))
input_matrix['ind_8',] <- c(as.numeric(NA), -0.2975866, as.numeric(NA), -0.18805090, -0.2293608, as.numeric(NA), as.numeric(NA))
input_matrix['ind_9',] <- c(as.numeric(NA), -0.2975866, as.numeric(NA), -0.18805090, -0.2293608, -0.3167715, as.numeric(NA))
input_grouping <- data.frame(matrix(data=c('ind_1','G1','ind_2','G1','ind_3','G1','ind_4','G2','ind_5','G2','ind_6','G2','ind_7','G3','ind_8','G4','ind_9','G5'), nrow=9, ncol=2, byrow=TRUE, dimnames = list(c(),c('ind','group'))), stringsAsFactors=FALSE)
# both ind are (#tp < df), trigger total mean curve (#tp < df)
input_matrix2 <- data.frame(matrix(nrow=2, ncol=7), stringsAsFactors = FALSE)
colnames(input_matrix2) <- c(0, 4, 8, 12, 24, 48, 72)
rownames(input_matrix2) <- c('ind_1','ind_2')
input_matrix2['ind_1',] <- c(-0.4101337, as.numeric(NA), 0.65175696, as.numeric(NA), as.numeric(NA), -0.2975552, -0.3135763)
input_matrix2['ind_2',] <- c(as.numeric(NA), -0.2975866, as.numeric(NA), -0.18805090, -0.2293608, -0.3167715, as.numeric(NA))
# no individual in the group 'dim(groupData.in)[1]==0' shouldn't be possible to trigger as matrix is only updated to data.frame if an individual is added
# no groupMean curve (group data tp < df ) shouldn't be possible to trigger as it's only checked if groupData.in is a data.frame (so minimum one ind)
## Expected
# with grouping
expected_obj <- list()
expected_obj$properties <- list()
expected_obj$properties$df <- 5
expected_obj$properties$CBand$status <- FALSE
expected_obj$properties$CBand$nBoot <- NA
expected_obj$properties$CBand$alpha <- NA
expected_obj$properties$pval.dist$status <- FALSE
expected_obj$properties$pval.dist$nPerm <- NA
expected_obj$properties$pval.dist$alpha <- NA
expected_obj$properties$pval.fit$status <- FALSE
expected_obj$properties$pval.fit$nPerm <- NA
expected_obj$properties$pval.fit$alpha <- NA
expected_obj$groups <- list()
expected_obj$groups$G1$rejectedInd <- list()
expected_obj$groups$G1$curveInd <- list()
expected_obj$groups$G1$curveInd$ind_1 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763), df=5)
expected_obj$groups$G1$curveInd$ind_2 <- stats::smooth.spline(c(0,4,8,12,24,48), c(-0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509), df=5)
expected_obj$groups$G1$curveInd$ind_3 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981), df=5)
expected_obj$groups$G1$groupMeanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.414391776666666656, 1.351222410000000096, 0.755240270000000047, 0.078953003333333313, -0.359661066666666696, -0.324312383333333343, -0.455263819999999986), df=7)
expected_obj$groups$G1$point.in <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,0,4,8,12,24,48,72, -0.41013370,3.77734620,0.65175696,-0.04287806,-0.27873830,-0.29755520,-0.31357630,-0.41013370,2.66841520,-0.08955276,-0.18796505,-0.31772180,-0.32895090,-0.41013370,2.49843700,-0.15771390,-0.24400003,-0.33381800,-0.33708020,-0.33309810), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G1$point.pred <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72, 0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G1$groupData.in <- input_matrix[1:3,]
expected_obj$groups$G1$groupData.pred <- data.frame(matrix(data=c(0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814), ncol=7, dimnames=list(c('ind_1','ind_2','ind_3'),c()), byrow=TRUE), stringsAsFactors=FALSE)
colnames(expected_obj$groups$G1$groupData.pred) <- colnames(input_matrix)
expected_obj$groups$G2$rejectedInd <- list()
expected_obj$groups$G2$curveInd <- list()
expected_obj$groups$G2$curveInd$ind_4 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3001967, -0.3649202, -0.2685244, -0.3401291, -0.3251096, -0.2837459), df=5)
expected_obj$groups$G2$curveInd$ind_5 <- stats::smooth.spline(c(0,4,8,12,48), c(-0.4101337, -0.3399216, -0.4101337, -0.3275887, -0.363232), df=5)
expected_obj$groups$G2$curveInd$ind_6 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968), df=5)
expected_obj$groups$G2$groupMeanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.39910886666666667, -0.35468173333333330, -0.36584719999999998, -0.32711873333333336, -0.27520023333333332, -0.36585533333333337, -0.44383700000000004), df=7)
expected_obj$groups$G2$point.in <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,48,0,4,8,12,24,48,72, -0.4101337,-0.3001967,-0.36492017,-0.26852436,-0.3401291,-0.3251096,-0.2837459,-0.4101337,-0.3399216,-0.41013374,-0.32758869,-0.3632320,-0.4101337,-0.3455376,-0.41013374,-0.33124182,-0.3650980,-0.4101337,-0.2877968), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G2$point.pred <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72, -0.3890455,-0.3450127,-0.3189128,-0.2994064,-0.3304448,-0.3262453,-0.283692,-0.4101337,-0.3399216,-0.4101337,-0.3275887,-0.1319989,-0.363232,-0.7591037,-0.3981474,-0.3791109,-0.3684951,-0.3543611,-0.363157,-0.4080887,-0.2887153), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj$groups$G2$groupData.in <- input_matrix[4:6,]
expected_obj$groups$G2$groupData.pred <- data.frame(matrix(data=c(-0.3890455,-0.3450127,-0.3189128,-0.2994064,-0.3304448,-0.3262453,-0.283692,-0.4101337,-0.3399216,-0.4101337,-0.3275887,-0.1319989,-0.363232,-0.7591037,-0.3981474,-0.3791109,-0.3684951,-0.3543611,-0.363157,-0.4080887,-0.2887153), ncol=7, dimnames=list(c('ind_4','ind_5','ind_6'),c()), byrow=TRUE), stringsAsFactors=FALSE)
colnames(expected_obj$groups$G2$groupData.pred) <- colnames(input_matrix)
expected_obj$groups$G3$rejectedInd <- list()
expected_obj$groups$G3$curveInd <- NA
expected_obj$groups$G3$groupMeanCurve <- NA
expected_obj$groups$G3$point.in <- NA
expected_obj$groups$G3$point.pred <- NA
expected_obj$groups$G3$groupData.in <- NA
expected_obj$groups$G3$groupData.pred <- NA
expected_obj$groups$G4$rejectedInd <- list()
expected_obj$groups$G4$rejectedInd$ind_8 <- input_matrix[8,]
expected_obj$groups$G4$curveInd <- NA
expected_obj$groups$G4$groupMeanCurve <- NA
expected_obj$groups$G4$point.in <- NA
expected_obj$groups$G4$point.pred <- NA
expected_obj$groups$G4$groupData.in <- NA
expected_obj$groups$G4$groupData.pred <- NA
expected_obj$groups$G5$rejectedInd <- list()
expected_obj$groups$G5$rejectedInd$ind_9 <- input_matrix[9,]
expected_obj$groups$G5$curveInd <- NA
expected_obj$groups$G5$groupMeanCurve <- NA
expected_obj$groups$G5$point.in <- NA
expected_obj$groups$G5$point.pred <- NA
expected_obj$groups$G5$groupData.in <- NA
expected_obj$groups$G5$groupData.pred <- NA
expected_obj$general <- list()
expected_obj$general$inputData <- input_matrix[-7,]
expected_obj$general$grouping <- input_grouping
expected_obj$general$cleanData.in <- input_matrix[1:6,]
expected_obj$general$cleanData.pred <- rbind(expected_obj$groups$G1$groupData.pred, expected_obj$groups$G2$groupData.pred)
expected_obj$general$pval.curveCorr <- NA
expected_obj$general$pval.dist <- NA
expected_obj$general$pval.dist.l <- NA
expected_obj$general$pval.dist.u <- NA
expected_obj$general$pval.fit <- NA
expected_obj$general$pval.fit.l <- NA
expected_obj$general$pval.fit.u <- NA
expected_obj$general$meanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.0076414549999999941, 0.4982703383333333402, 0.1946965350000000039, -0.1240828650000000144, -0.3174306499999999809, -0.3450838583333333265, -0.4495504100000000114), df=7)
# without grouping
expected_obj_noGrp <- list()
expected_obj_noGrp$properties <- list()
expected_obj_noGrp$properties$df <- 5
expected_obj_noGrp$properties$CBand$status <- FALSE
expected_obj_noGrp$properties$CBand$nBoot <- NA
expected_obj_noGrp$properties$CBand$alpha <- NA
expected_obj_noGrp$properties$pval.dist$status <- FALSE
expected_obj_noGrp$properties$pval.dist$nPerm <- NA
expected_obj_noGrp$properties$pval.dist$alpha <- NA
expected_obj_noGrp$properties$pval.fit$status <- FALSE
expected_obj_noGrp$properties$pval.fit$nPerm <- NA
expected_obj_noGrp$properties$pval.fit$alpha <- NA
expected_obj_noGrp$groups <- list()
expected_obj_noGrp$groups[[1]] <- list()
expected_obj_noGrp$groups[[1]]$rejectedInd <- list()
expected_obj_noGrp$groups[[1]]$rejectedInd[[7]] <- input_matrix[8,]
names(expected_obj_noGrp$groups[[1]]$rejectedInd)[[7]] <- 'ind_8'
expected_obj_noGrp$groups[[1]]$rejectedInd[[8]] <- input_matrix[9,]
names(expected_obj_noGrp$groups[[1]]$rejectedInd)[[8]] <- 'ind_9'
expected_obj_noGrp$groups[[1]]$curveInd <- list()
expected_obj_noGrp$groups[[1]]$curveInd$ind_1 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_2 <- stats::smooth.spline(c(0,4,8,12,24,48), c(-0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_3 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_4 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3001967, -0.3649202, -0.2685244, -0.3401291, -0.3251096, -0.2837459), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_5 <- stats::smooth.spline(c(0,4,8,12,48), c(-0.4101337, -0.3399216, -0.4101337, -0.3275887, -0.363232), df=5)
expected_obj_noGrp$groups[[1]]$curveInd$ind_6 <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(-0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968), df=5)
expected_obj_noGrp$groups[[1]]$groupMeanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.0076414616666666572, 0.4982703316666666771, 0.1946965333333333381, -0.1240828566666666716, -0.3174306383333333481, -0.3450838616666666581, -0.4495504083333333178), df=7)
expected_obj_noGrp$groups[[1]]$point.in <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,48,0,4,8,12,24,48,72, -0.4101337, 3.7773462, 0.65175696, -0.04287806, -0.2787383, -0.2975552, -0.3135763, -0.4101337, 2.6684152, -0.08955276, -0.18796505, -0.3177218, -0.3289509, -0.4101337, 2.4984370, -0.15771390, -0.24400003, -0.3338180, -0.3370802, -0.3330981, -0.4101337, -0.3001967, -0.3649202, -0.2685244, -0.3401291, -0.3251096, -0.2837459, -0.4101337, -0.3399216, -0.4101337, -0.3275887, -0.363232, -0.4101337, -0.3455376, -0.41013374, -0.33124182, -0.3650980, -0.4101337, -0.2877968), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj_noGrp$groups[[1]]$point.pred <- data.frame(matrix(data=c(0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72,0,4,8,12,24,48,72, 0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814,-0.38904549,-0.34501275,-0.3189128,-0.29940637,-0.33044479,-0.32624534,-0.28369199,-0.4101337,-0.3399216,-0.41013374,-0.32758869,-0.13199885,-0.363232,-0.75910373,-0.39814737,-0.37911089,-0.36849507,-0.35436109,-0.36315699,-0.40808868,-0.28871527), ncol=2, dimnames=list(c(),c('x','y')), byrow=FALSE), stringsAsFactors=FALSE)
expected_obj_noGrp$groups[[1]]$groupData.in <- input_matrix[1:6,]
expected_obj_noGrp$groups[[1]]$groupData.pred <- data.frame(matrix(data=c(0.79743448,1.55504776,1.21052367,0.49011151,-0.348835,-0.30607259,-0.31198823,0.01909793,1.67433373,0.56762228,-0.26372767,-0.33535016,-0.32788512,-0.72119509,0.42664292,0.82428574,0.48757486,0.01047517,-0.39479804,-0.33897944,-0.33260814,-0.38904549,-0.34501275,-0.3189128,-0.29940637,-0.33044479,-0.32624534,-0.28369199,-0.4101337,-0.3399216,-0.41013374,-0.32758869,-0.13199885,-0.363232,-0.75910373,-0.39814737,-0.37911089,-0.36849507,-0.35436109,-0.36315699,-0.40808868,-0.28871527), ncol=7, dimnames=list(c('ind_1','ind_2','ind_3','ind_4','ind_5','ind_6'),c()), byrow=TRUE), stringsAsFactors=FALSE)
colnames(expected_obj_noGrp$groups[[1]]$groupData.pred) <- colnames(input_matrix)
expected_obj_noGrp$general <- list()
expected_obj_noGrp$general$inputData <- input_matrix[-7,]
expected_obj_noGrp$general$grouping <- NA
expected_obj_noGrp$general$cleanData.in <- input_matrix[1:6,]
expected_obj_noGrp$general$cleanData.pred <- expected_obj_noGrp$groups[[1]]$groupData.pred
expected_obj_noGrp$general$pval.curveCorr <- NA
expected_obj_noGrp$general$pval.dist <- NA
expected_obj_noGrp$general$pval.dist.l <- NA
expected_obj_noGrp$general$pval.dist.u <- NA
expected_obj_noGrp$general$pval.fit <- NA
expected_obj_noGrp$general$pval.fit.l <- NA
expected_obj_noGrp$general$pval.fit.u <- NA
expected_obj_noGrp$general$meanCurve <- stats::smooth.spline(c(0,4,8,12,24,48,72), c(0.0076414616666666572, 0.4982703316666666771, 0.1946965333333333381, -0.1240828566666666716, -0.3174306383333333481, -0.3450838616666666581, -0.4495504083333333178), df=7)
# no total mean curve calculation
expected_obj_2 <- list()
expected_obj_2$properties <- list()
expected_obj_2$properties$df <- 5
expected_obj_2$properties$CBand$status <- FALSE
expected_obj_2$properties$CBand$nBoot <- NA
expected_obj_2$properties$CBand$alpha <- NA
expected_obj_2$properties$pval.dist$status <- FALSE
expected_obj_2$properties$pval.dist$nPerm <- NA
expected_obj_2$properties$pval.dist$alpha <- NA
expected_obj_2$properties$pval.fit$status <- FALSE
expected_obj_2$properties$pval.fit$nPerm <- NA
expected_obj_2$properties$pval.fit$alpha <- NA
expected_obj_2$groups <- list()
expected_obj_2$groups[[1]] <- list()
expected_obj_2$groups[[1]]$rejectedInd <- list()
expected_obj_2$groups[[1]]$rejectedInd[[1]] <- input_matrix2[1,]
names(expected_obj_2$groups[[1]]$rejectedInd)[[1]] <- 'ind_1'
expected_obj_2$groups[[1]]$rejectedInd[[2]] <- input_matrix2[2,]
names(expected_obj_2$groups[[1]]$rejectedInd)[[2]] <- 'ind_2'
expected_obj_2$groups[[1]]$curveInd <- NA
expected_obj_2$groups[[1]]$groupMeanCurve <- NA
expected_obj_2$groups[[1]]$point.in <- NA
expected_obj_2$groups[[1]]$point.pred <- NA
expected_obj_2$groups[[1]]$groupData.in <- NA
expected_obj_2$groups[[1]]$groupData.pred <- NA
expected_obj_2$general <- list()
expected_obj_2$general$inputData <- input_matrix2
expected_obj_2$general$grouping <- NA
expected_obj_2$general$cleanData.in <- matrix(ncol=7,nrow=0)
expected_obj_2$general$cleanData.pred <- matrix(ncol=7,nrow=0)
expected_obj_2$general$pval.curveCorr <- NA
expected_obj_2$general$pval.dist <- NA
expected_obj_2$general$pval.dist.l <- NA
expected_obj_2$general$pval.dist.u <- NA
expected_obj_2$general$pval.fit <- NA
expected_obj_2$general$pval.fit.l <- NA
expected_obj_2$general$pval.fit.u <- NA
expected_obj_2$general$meanCurve <- NA
test_that('default values, with grouping', {
# expected_fit
expected_fit <- expected_obj
# G1 G2 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
exp_curveInd_G1 <- expected_fit$groups$G1$curveInd
exp_curveInd_G2 <- expected_fit$groups$G2$curveInd
exp_grpMnCurve_G1 <- expected_fit$groups$G1$groupMeanCurve
exp_grpMnCurve_G2 <- expected_fit$groups$G2$groupMeanCurve
exp_meanCurve <- expected_fit$general$meanCurve
expected_fit$groups$G1$curveInd <- NULL
expected_fit$groups$G2$curveInd <- NULL
expected_fit$groups$G1$groupMeanCurve <- NULL
expected_fit$groups$G2$groupMeanCurve <- NULL
expected_fit$general$meanCurve <- NULL
# expected message
expected_message <- c("No individuals left in group #3 once data is imported (remove NA rows and columns)\n", "No individuals left in group #3\n", "ind_8 - #tp < 4\n", "No individuals left in group #4\n", "ind_9 - #tp(4) < df(5)\n", "No individuals left in group #5\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=input_grouping, verbose=TRUE))
result_fit_noV <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=input_grouping, verbose=FALSE))
# first check that verbose and no verbose match
expect_equal(result_fit$result, result_fit_noV$result)
# G1 G2 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
entries_to_check <- c("x","y","w","yin","tol","data","no.weights","lev","cv.crit","pen.crit","crit","df","spar","ratio","lambda","iparms","auxM","fit")
tryCatch({
tmp_curveInd_G1 <- result_fit$result$groups$G1$curveInd
tmp_curveInd_G2 <- result_fit$result$groups$G2$curveInd
tmp_grpMnCurve_G1 <- result_fit$result$groups$G1$groupMeanCurve
tmp_grpMnCurve_G2 <- result_fit$result$groups$G2$groupMeanCurve
tmp_meanCurve <- result_fit$result$general$meanCurve
result_fit$result$groups$G1$curveInd <- NULL
result_fit$result$groups$G2$curveInd <- NULL
result_fit$result$groups$G1$groupMeanCurve <- NULL
result_fit$result$groups$G2$groupMeanCurve <- NULL
result_fit$result$general$meanCurve <- NULL
},
error=function(cond) {
## catch, do nothing
tmp_curveInd_G1 <- NA
tmp_curveInd_G2 <- NA
tmp_grpMnCurve_G1 <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_grpMnCurve_G2 <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_meanCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
}
)
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check curveInd (without function call)
expect_equal(length(tmp_curveInd_G1), length(exp_curveInd_G1))
expect_equal(length(tmp_curveInd_G2), length(exp_curveInd_G2))
# curveInd G1
if (length(tmp_curveInd_G1) == length(exp_curveInd_G1)) {
for (i in seq_along(exp_curveInd_G1)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd_G1[[i]][[k]], exp_curveInd_G1[[i]][[k]], tolerance=1e-5)
}
}
}
# curveInd G2
if (length(tmp_curveInd_G2) == length(exp_curveInd_G2)) {
for (i in seq_along(exp_curveInd_G2)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd_G2[[i]][[k]], exp_curveInd_G2[[i]][[k]], tolerance=1e-5)
}
}
}
# Check groupMeanCurve (without function call)
# groupMeanCurve G1
if (!all(is.na(tmp_grpMnCurve_G1))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve_G1[[k]], exp_grpMnCurve_G1[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve_G1, exp_grpMnCurve_G1, tolerance=1e-5)
}
# groupMeanCurve G2
if (!all(is.na(tmp_grpMnCurve_G2))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve_G2[[k]], exp_grpMnCurve_G2[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve_G2, exp_grpMnCurve_G2, tolerance=1e-5)
}
# Check total meanCurve (without function call)
if (!all(is.na(tmp_meanCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_meanCurve[[k]], exp_meanCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_meanCurve, exp_meanCurve, tolerance=1e-5)
}
# Check result messages
expect_equal(length(result_fit$messages), 6)
expect_equal(result_fit$messages, expected_message)
# Check no message without verbose
expect_equal(length(result_fit_noV$messages), 0)
})
test_that('default values, no grouping', {
# expected_fit
expected_fit <- expected_obj_noGrp
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
exp_curveInd <- expected_fit$groups[[1]]$curveInd
exp_grpMnCurve <- expected_fit$groups[[1]]$groupMeanCurve
exp_meanCurve <- expected_fit$general$meanCurve
expected_fit$groups[[1]]$curveInd <- NULL
expected_fit$groups[[1]]$groupMeanCurve <- NULL
expected_fit$general$meanCurve <- NULL
# expected message
expected_message <- c("ind_8 - #tp < 4\n", "ind_9 - #tp(4) < df(5)\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=NA, verbose=TRUE))
result_fit_noV <- evaluate_promise(santaR_fit(input_matrix, df=5, grouping=NA, verbose=FALSE))
# first check that verbose and no verbose match
expect_equal(result_fit$result, result_fit_noV$result)
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
entries_to_check <- c("x","y","w","yin","tol","data","no.weights","lev","cv.crit","pen.crit","crit","df","spar","ratio","lambda","iparms","auxM","fit")
tryCatch({
tmp_curveInd <- result_fit$result$groups[[1]]$curveInd
tmp_grpMnCurve <- result_fit$result$groups[[1]]$groupMeanCurve
tmp_meanCurve <- result_fit$result$general$meanCurve
result_fit$result$groups[[1]]$curveInd <- NULL
result_fit$result$groups[[1]]$groupMeanCurve <- NULL
result_fit$result$general$meanCurve <- NULL
},
error=function(cond) {
## catch, do nothing
tmp_curveInd <- NA
tmp_grpMnCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_meanCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
}
)
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check curveInd (without function call)
expect_equal(length(tmp_curveInd), length(exp_curveInd))
# curveInd G1
if (length(tmp_curveInd) == length(exp_curveInd)) {
for (i in seq_along(exp_curveInd)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd[[i]][[k]], exp_curveInd[[i]][[k]], tolerance=1e-5)
}
}
}
# Check groupMeanCurve (without function call)
# groupMeanCurve G1
if (!all(is.na(tmp_grpMnCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve[[k]], exp_grpMnCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve, exp_grpMnCurve, tolerance=1e-5)
}
# Check total meanCurve (without function call)
if (!all(is.na(tmp_meanCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_meanCurve[[k]], exp_meanCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_meanCurve, exp_meanCurve, tolerance=1e-5)
}
# Check result messages
expect_equal(length(result_fit$messages), 2)
expect_equal(result_fit$messages, expected_message)
# Check no message without verbose
expect_equal(length(result_fit_noV$messages), 0)
})
test_that('no total mean curve calculation possible', {
# expected_fit
expected_fit <- expected_obj_2
# expected message
expected_message <- c("ind_1 - #tp(4) < df(5)\n", "ind_2 - #tp(4) < df(5)\n", "No individuals left in group #1\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(input_matrix2, df=5, grouping=NA, verbose=TRUE))
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check result messages
expect_equal(length(result_fit$messages), 3)
expect_equal(result_fit$messages, expected_message)
})
test_that('empty row and col in inputMatrix', {
# empty row (ind_7) & empty col (100) should be removed automatically (use expected_obj_noGrp)
# input
tmp_matrix <- cbind(input_matrix, "100"=c(NA,NA,NA,NA,NA,NA,NA,NA,NA))
# expected_fit
expected_fit <- expected_obj_noGrp
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
exp_curveInd <- expected_fit$groups[[1]]$curveInd
exp_grpMnCurve <- expected_fit$groups[[1]]$groupMeanCurve
exp_meanCurve <- expected_fit$general$meanCurve
expected_fit$groups[[1]]$curveInd <- NULL
expected_fit$groups[[1]]$groupMeanCurve <- NULL
expected_fit$general$meanCurve <- NULL
# expected message
expected_message <- c("ind_8 - #tp < 4\n", "ind_9 - #tp(4) < df(5)\n")
# results (output, warnings and messages)
result_fit <- evaluate_promise(santaR_fit(tmp_matrix, df=5, grouping=NA, verbose=TRUE))
result_fit_noV <- evaluate_promise(santaR_fit(tmp_matrix, df=5, grouping=NA, verbose=FALSE))
# first check that verbose and no verbose match
expect_equal(result_fit$result, result_fit_noV$result)
# G1 extract curveInd, groupMeanCurve + total meanCurve: smooth.spline function call cannot be compared
entries_to_check <- c("x","y","w","yin","tol","data","no.weights","lev","cv.crit","pen.crit","crit","df","spar","ratio","lambda","iparms","auxM","fit")
tryCatch({
tmp_curveInd <- result_fit$result$groups[[1]]$curveInd
tmp_grpMnCurve <- result_fit$result$groups[[1]]$groupMeanCurve
tmp_meanCurve <- result_fit$result$general$meanCurve
result_fit$result$groups[[1]]$curveInd <- NULL
result_fit$result$groups[[1]]$groupMeanCurve <- NULL
result_fit$result$general$meanCurve <- NULL
},
error=function(cond) {
## catch, do nothing
tmp_curveInd <- NA
tmp_grpMnCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
tmp_meanCurve <- list(x=NA,y=NA,w=NA,yin=NA,tol=NA,data=NA,no.weights=NA,lev=NA,cv.crit=NA,pen.crit=NA,crit=NA,df=NA,spar=NA,ratio=NA,lambda=NA,iparms=NA,auxM=NA,fit=NA)
}
)
# Check result (not the curveInd, groupMeanCurve, total meanCurve)
expect_equal(result_fit$result, expected_fit, tolerance=1e-5)
# Check curveInd (without function call)
expect_equal(length(tmp_curveInd), length(exp_curveInd))
# curveInd G1
if (length(tmp_curveInd) == length(exp_curveInd)) {
for (i in seq_along(exp_curveInd)) {
for (k in entries_to_check) {
expect_equal(tmp_curveInd[[i]][[k]], exp_curveInd[[i]][[k]], tolerance=1e-5)
}
}
}
# Check groupMeanCurve (without function call)
# groupMeanCurve G1
if (!all(is.na(tmp_grpMnCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_grpMnCurve[[k]], exp_grpMnCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_grpMnCurve, exp_grpMnCurve, tolerance=1e-5)
}
# Check total meanCurve (without function call)
if (!all(is.na(tmp_meanCurve))) {
for (k in entries_to_check) {
expect_equal(tmp_meanCurve[[k]], exp_meanCurve[[k]], tolerance=1e-5)
}
} else {
expect_equal(tmp_meanCurve, exp_meanCurve, tolerance=1e-5)
}
# Check result messages
expect_equal(length(result_fit$messages), 2)
expect_equal(result_fit$messages, expected_message)
# Check no message without verbose
expect_equal(length(result_fit_noV$messages), 0)
})
test_that('raise error', {
# df not numeric
expect_error(santaR_fit(input_matrix, df='notNumeric', grouping=input_grouping, verbose=FALSE), "Check input, 'df' should be numeric", fixed=TRUE)
# less than 4 tp
expect_error(santaR_fit(input_matrix[1:3], df=3, grouping=NA, verbose=FALSE), "Check input, a minimum of 4 unique time-points are required to fit a smooth.splines", fixed=TRUE)
# tp < df
expect_error(santaR_fit(input_matrix, df=10, grouping=NA, verbose=FALSE), "Check input, the number of unique time-points (7) cannot be less than the requested df (10)", fixed=TRUE)
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.