tests/testthat/test-select.R
In 16EAGLE/getSpatialData: Get different kinds of freely available spatial datasets
# -----------------------------------------------------------------------------------------
# DIRECTORIES
# -----------------------------------------------------------------------------------------
dir_records <- tt$resources$records
prio_products <- c(getSpatialData:::name_product_sentinel2(), getSpatialData:::name_product_landsat8(),
getSpatialData:::name_product_landsat7(), getSpatialData:::name_product_landsat5(),
getSpatialData:::name_product_landsatmss(), getSpatialData:::name_product_sentinel3(),
getSpatialData:::name_product_group_modis())
max_sub_period <- 10
min_distance <- 5
n_timestamps <- 3
# -----------------------------------------------------------------------------------------
# DEFINE TEST FUNCTIONS
# -----------------------------------------------------------------------------------------
# main test function
# ------------------------------------
clean_test_select <- function(dir_records, aoi_test, sensor, prio_products, tt, PREFIX, COLS, DATAFRAME, NUMERIC, CHARACTER) {
initialize_dir(tt$tmp)
modes <- c("select_unitemporal", "select_bitemporal", "select_timeseries")
max_sub_period <- 10
min_distance <- 5
n_timestamps <- 3
max_cloudcov_tile <- 80
records_read <- read_records(construct_filepath(dir_records, sensor, PREFIX$cmasks))
# calculate cloud masks
records <- calc_cloudcov(records_read, aoi = aoi_test, dir_out = tt$tmp)
# delete a cloud mask and test error
cloud_mask_files <- records[[getSpatialData:::name_cloud_mask_file()]]
unlink(cloud_mask_files[which(file.exists(cloud_mask_files))[1]])
expect_error(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
aoi = aoi_test, dir_out = tt$tmp))
preview_files <- records[[getSpatialData:::name_preview_file()]]
unlink(preview_files[which(file.exists(preview_files))[1]])
expect_error(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
aoi = aoi_test, dir_out = tt$tmp))
records <- calc_cloudcov(records_read, aoi = aoi_test, dir_out = tt$tmp)
for (mode in modes) {
is_unitemporal <- grepl("unitemporal", mode)
is_bitemporal <- grepl("bitemporal", mode)
is_timeseries <- grepl("timeseries", mode)
as_sf <- sample(c(TRUE, FALSE), 1)
expected_class <- ifelse(as_sf, SF, DATAFRAME)
if (is_unitemporal) {
test_unitemporal_without_writing(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test)
records_unitemporal <- expect_is(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
aoi = aoi_test, dir_out = tt$tmp, as_sf = as_sf), expected_class)
records_select <- records_unitemporal
} else if (is_bitemporal) {
test_bitemporal_without_writing(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test)
records_bitemporal <- expect_is(select_bitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
min_distance = min_distance,
aoi = aoi_test, dir_out = tt$tmp, as_sf = as_sf), expected_class)
records_select <- records_bitemporal
} else if (is_timeseries) {
test_timeseries_without_writing(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test)
records_timeseries <- expect_is(select_timeseries(records,
n_timestamps = n_timestamps,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
min_distance = min_distance,
aoi = aoi_test, dir_out = tt$tmp, as_sf = as_sf), expected_class)
records_select <- records_timeseries
}
cols_given <- names(records_select)
# check if column exists
expect_true(COLS$pmos_col %in% cols_given)
expect_true(COLS$cmos_col %in% cols_given)
expect_true(COLS$timestamp_col %in% cols_given)
# get column vectors
cmos_col <- records_select[[COLS$cmos_col]]
pmos_col <- records_select[[COLS$pmos_col]]
timestamp_col <- records_select[[COLS$timestamp_col]]
# check column data type
expect_is(pmos_col, CHARACTER)
expect_is(cmos_col, CHARACTER)
expect_true(any(inherits(timestamp_col, NUMERIC), inherits(timestamp_col, INTEGER)))
# check rasters
for (file in cmos_col[!is.na(cmos_col)]) {
expect_true(file.exists(file))
loaded_cmos <- test_raster_read(file)
expect_false(is.na(crs(loaded_cmos)) && is.na(st_crs(loaded_cmos)))
# values
expect_equal(minValue(loaded_cmos), 1) # not 0!
expect_equal(maxValue(loaded_cmos), 1)
}
for (file in pmos_col[!is.na(pmos_col)]) {
expect_true(file.exists(file))
loaded_pmos <- test_stack_read(file)
expect_false(is.na(crs(loaded_pmos)) && is.na(st_crs(loaded_pmos)))
}
is_selected <- !is.na(timestamp_col)
# check if max_cloudcov_tile is fulfilled
cloudcov <- records_select[is_selected,][[getSpatialData:::name_cloudcov()]]
cloudcov_filtered <- as.numeric(cloudcov[!is.na(cloudcov)])
if (length(cloudcov_filtered) > 0) {
expect_true(all(cloudcov_filtered < max_cloudcov_tile))
}
# check if temporal args are fulfilled
timestamp_col_filtered <- timestamp_col[is_selected]
dates <- sapply(records_select[is_selected,][[getSpatialData:::name_date_acquisition()]], as.Date)
if (is_unitemporal) {
expect_true(unique(timestamp_col_filtered) == 1)
if (length(dates) > 1) {
start_date <- min(dates)
end_date <- max(dates)
sub_period_length <- end_date - start_date + 1
expect_true(sub_period_length <= max_sub_period)
}
} else if (is_bitemporal) {
expect_true(all(timestamp_col_filtered < 3))
}
if (is_bitemporal || is_timeseries) {
previous_last_date <- 0
for (ts in sort(unique(timestamp_col_filtered))) {
match <- which(records_select[[COLS$timestamp_col]] == ts)
dates_matched <- as.Date(records_select[match,][[getSpatialData:::name_date_acquisition()]])
start_date <- min(dates_matched)
end_date <- max(dates_matched)
sub_period_length <- as.numeric(as.Date(end_date) - as.Date(start_date)) + 1
expect_true(sub_period_length <= max_sub_period)
if (ts > 1) {
expect_true((start_date - previous_last_date) > min_distance)
}
previous_last_date <- end_date
}
}
}
finish_dir(tt$tmp)
}
# test without writing preview mosaic respectively cloud mask mosaic
validate_output <- function(records, column_name_given, column_name_not_given) {
names <- names(records)
expect_false(column_name_not_given %in% names)
expect_true(column_name_given %in% names)
not_na <- !is.na(records[[column_name_given]])
if (any(not_na)) {
expect_true(all(file.exists(records[[column_name_given]][which(not_na)])))
}
}
# unitemporal
test_unitemporal_without_writing <- function(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test) {
records_without_cmos <- expect_is(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_cmask_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_cmos, COLS$pmos_col, COLS$cmos_col)
records_without_pmos <- expect_is(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_preview_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_pmos, COLS$cmos_col, COLS$pmos_col)
}
# bitemporal
test_bitemporal_without_writing <- function(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test) {
records_without_cmos <- expect_is(select_bitemporal(records,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_cmask_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_cmos, COLS$pmos_col, COLS$cmos_col)
records_without_pmos <- expect_is(select_bitemporal(records,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_preview_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_pmos, COLS$cmos_col, COLS$pmos_col)
}
# timeseries
test_timeseries_without_writing <- function(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test) {
records_without_cmos <- expect_is(select_timeseries(records,
n_timestamps = 3,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_cmask_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_cmos, COLS$pmos_col, COLS$cmos_col)
records_without_pmos <- expect_is(select_timeseries(records,
n_timestamps = 3,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_preview_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_pmos, COLS$cmos_col, COLS$pmos_col)
}
# tests errors
# ------------------------------------
error_test_select <- function(records, aoi, tt) {
# records type
flawed_records <- "Dumbledore"
max_sub <- 10
n_timestamps = 3
min_dist = 6
expect_error(select_unitemporal(records = flawed_records, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), RECORDS_TYPE_ERROR)
expect_error(select_bitemporal(records = flawed_records, max_sub_period = max_sub, min_distance = min_dist,
aoi = aoi, dir_out = tt$tmp), RECORDS_TYPE_ERROR)
expect_error(select_timeseries(records = flawed_records, n_timestamps = n_timestamps, max_sub_period = max_sub,
min_distance = min_dist, aoi = aoi, dir_out = tt$tmp), RECORDS_TYPE_ERROR)
# records column missing
needed_cols <- getSpatialData:::.get_needed_cols_select()
for (col_remove in needed_cols) {
input1_records <- records
input1_records[[col_remove]] <- NULL
expect_error(select_unitemporal(input1_records,
max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), column_error_msg(col_remove))
expect_error(select_bitemporal(input1_records,
max_sub_period = max_sub,
min_distance = min_dist,
aoi = aoi, dir_out = tt$tmp), column_error_msg(col_remove))
expect_error(select_timeseries(input1_records,
n_timestamps = n_timestamps,
max_sub_period = max_sub,
min_distance = min_dist,
aoi = aoi, dir_out = tt$tmp), column_error_msg(col_remove))
}
# max_sub_period
max_sub_period_name <- "max_sub_period"
input1_max_sub_period <- "Nagini"
error1_max_sub_period <- type_error_msg(input1_max_sub_period, max_sub_period_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = input1_max_sub_period,
aoi = aoi, dir_out = tt$tmp), error1_max_sub_period) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = input1_max_sub_period,
aoi = aoi, dir_out = tt$tmp), error1_max_sub_period) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = input1_max_sub_period,
aoi = aoi, dir_out = tt$tmp), error1_max_sub_period) # TS
# min_distance
min_distance_name <- "min_distance"
input1_min_distance <- "Malfoy"
error1_min_distance <- type_error_msg(input1_min_distance, min_distance_name, NUMERIC)
expect_error(select_bitemporal(records, min_distance = input1_min_distance, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), error1_min_distance) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = input1_min_distance, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), error1_min_distance) # TS
# n_timestamps
num_timestamps_name <- "n_timestamps"
input1_num_timestamps <- "Lucius"
error1_num_timestamps <- type_error_msg(input1_num_timestamps, num_timestamps_name, NUMERIC)
expect_error(select_timeseries(records, n_timestamps = input1_num_timestamps,
min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), error1_num_timestamps) # TS
# min_improvement
min_improvement_name <- "min_improvement"
input1_min_improvement <- "Dolores"
error1_min_improvment <- type_error_msg(input1_min_improvement, min_improvement_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
min_improvement = input1_min_improvement,
aoi = aoi, dir_out = tt$tmp), error1_min_improvment) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
min_improvement = input1_min_improvement,
aoi = aoi, dir_out = tt$tmp), error1_min_improvment) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
min_improvement = input1_min_improvement,
aoi = aoi, dir_out = tt$tmp), error1_min_improvment) # TS
# max_cloudcov_tile
max_cloudcov_tile_name <- "max_cloudcov_tile"
input1_max_cloudcov_tile <- "Umbridge"
error1_max_cloudcov_tile <- type_error_msg(input1_max_cloudcov_tile, max_cloudcov_tile_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
max_cloudcov_tile = input1_max_cloudcov_tile,
aoi = aoi, dir_out = tt$tmp), error1_max_cloudcov_tile) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
max_cloudcov_tile = input1_max_cloudcov_tile,
aoi = aoi, dir_out = tt$tmp), error1_max_cloudcov_tile) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
max_cloudcov_tile = input1_max_cloudcov_tile,
aoi = aoi, dir_out = tt$tmp), error1_max_cloudcov_tile) # TS
# satisfaction_value
satisfaction_value_name <- "satisfaction_value"
input1_satisfaction_value <- "Wormtail"
error1_satisfaction_value <- type_error_msg(input1_satisfaction_value, satisfaction_value_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
satisfaction_value = input1_satisfaction_value,
aoi = aoi, dir_out = tt$tmp), error1_satisfaction_value) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
satisfaction_value = input1_satisfaction_value,
aoi = aoi, dir_out = tt$tmp), error1_satisfaction_value) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
satisfaction_value = input1_satisfaction_value,
aoi = aoi, dir_out = tt$tmp), error1_satisfaction_value) # TS
# prio_products
prio_products_name <- "prio_products"
input1_prio_products <- c(100, 200)
error1_prio_products <- type_error_msg(input1_prio_products, prio_products_name, CHARACTER)
input2_prio_products <- c("Sentinel-2", "Kedavra")
error2_prio_products <- "Argument 'prio_products' has to be provided with sensor names in the same format as returned by get_select_supported()"
expect_error(select_unitemporal(records, max_sub_period = max_sub,
prio_products = input1_prio_products,
aoi = aoi, dir_out = tt$tmp), error1_prio_products) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input1_prio_products,
aoi = aoi, dir_out = tt$tmp), error1_prio_products)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input1_prio_products,
aoi = aoi, dir_out = tt$tmp), error1_prio_products)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
prio_products = input2_prio_products,
aoi = aoi, dir_out = tt$tmp), error2_prio_products) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input2_prio_products,
aoi = aoi, dir_out = tt$tmp), error2_prio_products)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input2_prio_products,
aoi = aoi, dir_out = tt$tmp), error2_prio_products)
# aoi
input1_aoi <- NULL
input2_aoi <- "Crucio"
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = input1_aoi, dir_out = tt$tmp)) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = input1_aoi, dir_out = tt$tmp))
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = input1_aoi, dir_out = tt$tmp))
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = input2_aoi, dir_out = tt$tmp)) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = input2_aoi, dir_out = tt$tmp))
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = input2_aoi, dir_out = tt$tmp))
# dir_out
input1_dir_out <- 10
error1_dir_out <- type_error_msg(input1_dir_out, "dir_out", CHARACTER)
input2_dir_out <- "Quirrell"
error2_dir_out <- dir_out_error_msg(input2_dir_out)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = aoi, dir_out = input1_dir_out), error1_dir_out) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input1_dir_out), error1_dir_out)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input1_dir_out), error1_dir_out)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = aoi, dir_out = input2_dir_out), error2_dir_out) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input2_dir_out), error2_dir_out)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input2_dir_out), error2_dir_out)
# verbose
input1_verbose <- "of course"
error1_verbose <- type_error_msg(input1_verbose, "verbose", LOGICAL)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp, verbose = input1_verbose), error1_verbose) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp, verbose = input1_verbose), error1_verbose)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp, verbose = input1_verbose), error1_verbose)
}
# -----------------------------------------------------------------------------------------
# RUN TESTS
# -----------------------------------------------------------------------------------------
# errors
# ------------------------------------
# TEST 1
# -------
# Target: Test errors
records <- read_records(construct_filepath(dir_records, SENTINEL2, PREFIX$cmasks))
# with false input
error_test_select(records, aoi = aoi_test, tt)
# clean input
# ------------------------------------
# TEST 2
# -------
# Target: test with Sentinel-1
initialize_dir(tt$tmp)
records <- read_records(construct_filepath(dir_records, SENTINEL1, PREFIX$cmasks))
# Target: test unitemporal with Sentinel-1
records_select <- select_unitemporal(records, max_sub_period = max_sub_period, aoi = aoi_test, dir_out = tt$tmp)
finish_dir(tt$tmp)
initialize_dir(tt$tmp)
# Target: test bitemporal with Sentinel-1
records_select <- select_bitemporal(records, min_distance = min_distance, max_sub_period = max_sub_period, aoi = aoi_test,, dir_out = tt$tmp)
finish_dir(tt$tmp)
initialize_dir(tt$tmp)
# Target: test timeseries with Sentinel-1
records_select <- select_timeseries(records,
n_timestamps = n_timestamps,
min_distance = min_distance,
max_sub_period = max_sub_period,
aoi = aoi_test,
dir_out = tt$tmp)
finish_dir(tt$tmp)
# TEST 3
# -------
# Target: test with Sentinel-2
clean_test_select(dir_records, aoi_test = aoi_test, sensor = SENTINEL2,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 4
# -------
# Target: test with Sentinel-3
clean_test_select(dir_records, aoi_test = aoi_test, sensor = SENTINEL3,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 5
# -------
# Target: test with Landsat
clean_test_select(dir_records, aoi_test = aoi_test, sensor = LANDSAT,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 6
# -------
# Target: test with MODIS
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MODIS,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 7
# -------
# Target: test with mixed sensors without prio_products
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MIXED,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 8
# -------
# Target: test with mixed sensors with prio_products
prio_products <- c(getSpatialData:::name_product_sentinel2(),
getSpatialData:::name_product_group_modis(),
getSpatialData:::name_product_group_landsat())
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MIXED,
prio_products = prio_products, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 9
# -------
# Target: test with mixed sensors with prio_products and Sentinel-1
prio_products <- c(getSpatialData:::name_product_sentinel2(), getSpatialData:::name_product_landsat7())
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MIXED,
prio_products = prio_products, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
16EAGLE/getSpatialData documentation built on June 9, 2022, 11:28 a.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.
# -----------------------------------------------------------------------------------------
# DIRECTORIES
# -----------------------------------------------------------------------------------------
dir_records <- tt$resources$records
prio_products <- c(getSpatialData:::name_product_sentinel2(), getSpatialData:::name_product_landsat8(),
getSpatialData:::name_product_landsat7(), getSpatialData:::name_product_landsat5(),
getSpatialData:::name_product_landsatmss(), getSpatialData:::name_product_sentinel3(),
getSpatialData:::name_product_group_modis())
max_sub_period <- 10
min_distance <- 5
n_timestamps <- 3
# -----------------------------------------------------------------------------------------
# DEFINE TEST FUNCTIONS
# -----------------------------------------------------------------------------------------
# main test function
# ------------------------------------
clean_test_select <- function(dir_records, aoi_test, sensor, prio_products, tt, PREFIX, COLS, DATAFRAME, NUMERIC, CHARACTER) {
initialize_dir(tt$tmp)
modes <- c("select_unitemporal", "select_bitemporal", "select_timeseries")
max_sub_period <- 10
min_distance <- 5
n_timestamps <- 3
max_cloudcov_tile <- 80
records_read <- read_records(construct_filepath(dir_records, sensor, PREFIX$cmasks))
# calculate cloud masks
records <- calc_cloudcov(records_read, aoi = aoi_test, dir_out = tt$tmp)
# delete a cloud mask and test error
cloud_mask_files <- records[[getSpatialData:::name_cloud_mask_file()]]
unlink(cloud_mask_files[which(file.exists(cloud_mask_files))[1]])
expect_error(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
aoi = aoi_test, dir_out = tt$tmp))
preview_files <- records[[getSpatialData:::name_preview_file()]]
unlink(preview_files[which(file.exists(preview_files))[1]])
expect_error(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
aoi = aoi_test, dir_out = tt$tmp))
records <- calc_cloudcov(records_read, aoi = aoi_test, dir_out = tt$tmp)
for (mode in modes) {
is_unitemporal <- grepl("unitemporal", mode)
is_bitemporal <- grepl("bitemporal", mode)
is_timeseries <- grepl("timeseries", mode)
as_sf <- sample(c(TRUE, FALSE), 1)
expected_class <- ifelse(as_sf, SF, DATAFRAME)
if (is_unitemporal) {
test_unitemporal_without_writing(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test)
records_unitemporal <- expect_is(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
aoi = aoi_test, dir_out = tt$tmp, as_sf = as_sf), expected_class)
records_select <- records_unitemporal
} else if (is_bitemporal) {
test_bitemporal_without_writing(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test)
records_bitemporal <- expect_is(select_bitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
min_distance = min_distance,
aoi = aoi_test, dir_out = tt$tmp, as_sf = as_sf), expected_class)
records_select <- records_bitemporal
} else if (is_timeseries) {
test_timeseries_without_writing(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test)
records_timeseries <- expect_is(select_timeseries(records,
n_timestamps = n_timestamps,
max_cloudcov_tile = max_cloudcov_tile,
prio_products = prio_products,
max_sub_period = max_sub_period,
min_distance = min_distance,
aoi = aoi_test, dir_out = tt$tmp, as_sf = as_sf), expected_class)
records_select <- records_timeseries
}
cols_given <- names(records_select)
# check if column exists
expect_true(COLS$pmos_col %in% cols_given)
expect_true(COLS$cmos_col %in% cols_given)
expect_true(COLS$timestamp_col %in% cols_given)
# get column vectors
cmos_col <- records_select[[COLS$cmos_col]]
pmos_col <- records_select[[COLS$pmos_col]]
timestamp_col <- records_select[[COLS$timestamp_col]]
# check column data type
expect_is(pmos_col, CHARACTER)
expect_is(cmos_col, CHARACTER)
expect_true(any(inherits(timestamp_col, NUMERIC), inherits(timestamp_col, INTEGER)))
# check rasters
for (file in cmos_col[!is.na(cmos_col)]) {
expect_true(file.exists(file))
loaded_cmos <- test_raster_read(file)
expect_false(is.na(crs(loaded_cmos)) && is.na(st_crs(loaded_cmos)))
# values
expect_equal(minValue(loaded_cmos), 1) # not 0!
expect_equal(maxValue(loaded_cmos), 1)
}
for (file in pmos_col[!is.na(pmos_col)]) {
expect_true(file.exists(file))
loaded_pmos <- test_stack_read(file)
expect_false(is.na(crs(loaded_pmos)) && is.na(st_crs(loaded_pmos)))
}
is_selected <- !is.na(timestamp_col)
# check if max_cloudcov_tile is fulfilled
cloudcov <- records_select[is_selected,][[getSpatialData:::name_cloudcov()]]
cloudcov_filtered <- as.numeric(cloudcov[!is.na(cloudcov)])
if (length(cloudcov_filtered) > 0) {
expect_true(all(cloudcov_filtered < max_cloudcov_tile))
}
# check if temporal args are fulfilled
timestamp_col_filtered <- timestamp_col[is_selected]
dates <- sapply(records_select[is_selected,][[getSpatialData:::name_date_acquisition()]], as.Date)
if (is_unitemporal) {
expect_true(unique(timestamp_col_filtered) == 1)
if (length(dates) > 1) {
start_date <- min(dates)
end_date <- max(dates)
sub_period_length <- end_date - start_date + 1
expect_true(sub_period_length <= max_sub_period)
}
} else if (is_bitemporal) {
expect_true(all(timestamp_col_filtered < 3))
}
if (is_bitemporal || is_timeseries) {
previous_last_date <- 0
for (ts in sort(unique(timestamp_col_filtered))) {
match <- which(records_select[[COLS$timestamp_col]] == ts)
dates_matched <- as.Date(records_select[match,][[getSpatialData:::name_date_acquisition()]])
start_date <- min(dates_matched)
end_date <- max(dates_matched)
sub_period_length <- as.numeric(as.Date(end_date) - as.Date(start_date)) + 1
expect_true(sub_period_length <= max_sub_period)
if (ts > 1) {
expect_true((start_date - previous_last_date) > min_distance)
}
previous_last_date <- end_date
}
}
}
finish_dir(tt$tmp)
}
# test without writing preview mosaic respectively cloud mask mosaic
validate_output <- function(records, column_name_given, column_name_not_given) {
names <- names(records)
expect_false(column_name_not_given %in% names)
expect_true(column_name_given %in% names)
not_na <- !is.na(records[[column_name_given]])
if (any(not_na)) {
expect_true(all(file.exists(records[[column_name_given]][which(not_na)])))
}
}
# unitemporal
test_unitemporal_without_writing <- function(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test) {
records_without_cmos <- expect_is(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_cmask_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_cmos, COLS$pmos_col, COLS$cmos_col)
records_without_pmos <- expect_is(select_unitemporal(records,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_preview_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_pmos, COLS$cmos_col, COLS$pmos_col)
}
# bitemporal
test_bitemporal_without_writing <- function(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test) {
records_without_cmos <- expect_is(select_bitemporal(records,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_cmask_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_cmos, COLS$pmos_col, COLS$cmos_col)
records_without_pmos <- expect_is(select_bitemporal(records,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_preview_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_pmos, COLS$cmos_col, COLS$pmos_col)
}
# timeseries
test_timeseries_without_writing <- function(records, expected_class, max_cloudcov_tile, as_sf, tt, COLS, aoi_test) {
records_without_cmos <- expect_is(select_timeseries(records,
n_timestamps = 3,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_cmask_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_cmos, COLS$pmos_col, COLS$cmos_col)
records_without_pmos <- expect_is(select_timeseries(records,
n_timestamps = 3,
min_distance = 5,
max_cloudcov_tile = max_cloudcov_tile,
max_sub_period = 25,
aoi = aoi_test,
write_preview_mosaic = FALSE,
dir_out = tt$tmp, as_sf = as_sf), expected_class)
validate_output(records_without_pmos, COLS$cmos_col, COLS$pmos_col)
}
# tests errors
# ------------------------------------
error_test_select <- function(records, aoi, tt) {
# records type
flawed_records <- "Dumbledore"
max_sub <- 10
n_timestamps = 3
min_dist = 6
expect_error(select_unitemporal(records = flawed_records, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), RECORDS_TYPE_ERROR)
expect_error(select_bitemporal(records = flawed_records, max_sub_period = max_sub, min_distance = min_dist,
aoi = aoi, dir_out = tt$tmp), RECORDS_TYPE_ERROR)
expect_error(select_timeseries(records = flawed_records, n_timestamps = n_timestamps, max_sub_period = max_sub,
min_distance = min_dist, aoi = aoi, dir_out = tt$tmp), RECORDS_TYPE_ERROR)
# records column missing
needed_cols <- getSpatialData:::.get_needed_cols_select()
for (col_remove in needed_cols) {
input1_records <- records
input1_records[[col_remove]] <- NULL
expect_error(select_unitemporal(input1_records,
max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), column_error_msg(col_remove))
expect_error(select_bitemporal(input1_records,
max_sub_period = max_sub,
min_distance = min_dist,
aoi = aoi, dir_out = tt$tmp), column_error_msg(col_remove))
expect_error(select_timeseries(input1_records,
n_timestamps = n_timestamps,
max_sub_period = max_sub,
min_distance = min_dist,
aoi = aoi, dir_out = tt$tmp), column_error_msg(col_remove))
}
# max_sub_period
max_sub_period_name <- "max_sub_period"
input1_max_sub_period <- "Nagini"
error1_max_sub_period <- type_error_msg(input1_max_sub_period, max_sub_period_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = input1_max_sub_period,
aoi = aoi, dir_out = tt$tmp), error1_max_sub_period) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = input1_max_sub_period,
aoi = aoi, dir_out = tt$tmp), error1_max_sub_period) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = input1_max_sub_period,
aoi = aoi, dir_out = tt$tmp), error1_max_sub_period) # TS
# min_distance
min_distance_name <- "min_distance"
input1_min_distance <- "Malfoy"
error1_min_distance <- type_error_msg(input1_min_distance, min_distance_name, NUMERIC)
expect_error(select_bitemporal(records, min_distance = input1_min_distance, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), error1_min_distance) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = input1_min_distance, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), error1_min_distance) # TS
# n_timestamps
num_timestamps_name <- "n_timestamps"
input1_num_timestamps <- "Lucius"
error1_num_timestamps <- type_error_msg(input1_num_timestamps, num_timestamps_name, NUMERIC)
expect_error(select_timeseries(records, n_timestamps = input1_num_timestamps,
min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp), error1_num_timestamps) # TS
# min_improvement
min_improvement_name <- "min_improvement"
input1_min_improvement <- "Dolores"
error1_min_improvment <- type_error_msg(input1_min_improvement, min_improvement_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
min_improvement = input1_min_improvement,
aoi = aoi, dir_out = tt$tmp), error1_min_improvment) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
min_improvement = input1_min_improvement,
aoi = aoi, dir_out = tt$tmp), error1_min_improvment) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
min_improvement = input1_min_improvement,
aoi = aoi, dir_out = tt$tmp), error1_min_improvment) # TS
# max_cloudcov_tile
max_cloudcov_tile_name <- "max_cloudcov_tile"
input1_max_cloudcov_tile <- "Umbridge"
error1_max_cloudcov_tile <- type_error_msg(input1_max_cloudcov_tile, max_cloudcov_tile_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
max_cloudcov_tile = input1_max_cloudcov_tile,
aoi = aoi, dir_out = tt$tmp), error1_max_cloudcov_tile) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
max_cloudcov_tile = input1_max_cloudcov_tile,
aoi = aoi, dir_out = tt$tmp), error1_max_cloudcov_tile) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
max_cloudcov_tile = input1_max_cloudcov_tile,
aoi = aoi, dir_out = tt$tmp), error1_max_cloudcov_tile) # TS
# satisfaction_value
satisfaction_value_name <- "satisfaction_value"
input1_satisfaction_value <- "Wormtail"
error1_satisfaction_value <- type_error_msg(input1_satisfaction_value, satisfaction_value_name, NUMERIC)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
satisfaction_value = input1_satisfaction_value,
aoi = aoi, dir_out = tt$tmp), error1_satisfaction_value) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
satisfaction_value = input1_satisfaction_value,
aoi = aoi, dir_out = tt$tmp), error1_satisfaction_value) # BT
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
satisfaction_value = input1_satisfaction_value,
aoi = aoi, dir_out = tt$tmp), error1_satisfaction_value) # TS
# prio_products
prio_products_name <- "prio_products"
input1_prio_products <- c(100, 200)
error1_prio_products <- type_error_msg(input1_prio_products, prio_products_name, CHARACTER)
input2_prio_products <- c("Sentinel-2", "Kedavra")
error2_prio_products <- "Argument 'prio_products' has to be provided with sensor names in the same format as returned by get_select_supported()"
expect_error(select_unitemporal(records, max_sub_period = max_sub,
prio_products = input1_prio_products,
aoi = aoi, dir_out = tt$tmp), error1_prio_products) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input1_prio_products,
aoi = aoi, dir_out = tt$tmp), error1_prio_products)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input1_prio_products,
aoi = aoi, dir_out = tt$tmp), error1_prio_products)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
prio_products = input2_prio_products,
aoi = aoi, dir_out = tt$tmp), error2_prio_products) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input2_prio_products,
aoi = aoi, dir_out = tt$tmp), error2_prio_products)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
prio_products = input2_prio_products,
aoi = aoi, dir_out = tt$tmp), error2_prio_products)
# aoi
input1_aoi <- NULL
input2_aoi <- "Crucio"
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = input1_aoi, dir_out = tt$tmp)) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = input1_aoi, dir_out = tt$tmp))
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = input1_aoi, dir_out = tt$tmp))
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = input2_aoi, dir_out = tt$tmp)) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = input2_aoi, dir_out = tt$tmp))
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = input2_aoi, dir_out = tt$tmp))
# dir_out
input1_dir_out <- 10
error1_dir_out <- type_error_msg(input1_dir_out, "dir_out", CHARACTER)
input2_dir_out <- "Quirrell"
error2_dir_out <- dir_out_error_msg(input2_dir_out)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = aoi, dir_out = input1_dir_out), error1_dir_out) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input1_dir_out), error1_dir_out)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input1_dir_out), error1_dir_out)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = aoi, dir_out = input2_dir_out), error2_dir_out) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input2_dir_out), error2_dir_out)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = input2_dir_out), error2_dir_out)
# verbose
input1_verbose <- "of course"
error1_verbose <- type_error_msg(input1_verbose, "verbose", LOGICAL)
expect_error(select_unitemporal(records, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp, verbose = input1_verbose), error1_verbose) # UT
expect_error(select_bitemporal(records, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp, verbose = input1_verbose), error1_verbose)
expect_error(select_timeseries(records, n_timestamps = n_timestamps, min_distance = min_dist, max_sub_period = max_sub,
aoi = aoi, dir_out = tt$tmp, verbose = input1_verbose), error1_verbose)
}
# -----------------------------------------------------------------------------------------
# RUN TESTS
# -----------------------------------------------------------------------------------------
# errors
# ------------------------------------
# TEST 1
# -------
# Target: Test errors
records <- read_records(construct_filepath(dir_records, SENTINEL2, PREFIX$cmasks))
# with false input
error_test_select(records, aoi = aoi_test, tt)
# clean input
# ------------------------------------
# TEST 2
# -------
# Target: test with Sentinel-1
initialize_dir(tt$tmp)
records <- read_records(construct_filepath(dir_records, SENTINEL1, PREFIX$cmasks))
# Target: test unitemporal with Sentinel-1
records_select <- select_unitemporal(records, max_sub_period = max_sub_period, aoi = aoi_test, dir_out = tt$tmp)
finish_dir(tt$tmp)
initialize_dir(tt$tmp)
# Target: test bitemporal with Sentinel-1
records_select <- select_bitemporal(records, min_distance = min_distance, max_sub_period = max_sub_period, aoi = aoi_test,, dir_out = tt$tmp)
finish_dir(tt$tmp)
initialize_dir(tt$tmp)
# Target: test timeseries with Sentinel-1
records_select <- select_timeseries(records,
n_timestamps = n_timestamps,
min_distance = min_distance,
max_sub_period = max_sub_period,
aoi = aoi_test,
dir_out = tt$tmp)
finish_dir(tt$tmp)
# TEST 3
# -------
# Target: test with Sentinel-2
clean_test_select(dir_records, aoi_test = aoi_test, sensor = SENTINEL2,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 4
# -------
# Target: test with Sentinel-3
clean_test_select(dir_records, aoi_test = aoi_test, sensor = SENTINEL3,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 5
# -------
# Target: test with Landsat
clean_test_select(dir_records, aoi_test = aoi_test, sensor = LANDSAT,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 6
# -------
# Target: test with MODIS
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MODIS,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 7
# -------
# Target: test with mixed sensors without prio_products
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MIXED,
prio_products = NULL, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 8
# -------
# Target: test with mixed sensors with prio_products
prio_products <- c(getSpatialData:::name_product_sentinel2(),
getSpatialData:::name_product_group_modis(),
getSpatialData:::name_product_group_landsat())
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MIXED,
prio_products = prio_products, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
# TEST 9
# -------
# Target: test with mixed sensors with prio_products and Sentinel-1
prio_products <- c(getSpatialData:::name_product_sentinel2(), getSpatialData:::name_product_landsat7())
clean_test_select(dir_records, aoi_test = aoi_test, sensor = MIXED,
prio_products = prio_products, tt = tt, PREFIX = PREFIX, COLS = COLS,
DATAFRAME = DATAFRAME, NUMERIC = NUMERIC, CHARACTER = CHARACTER)
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.