22 |
23 | Package website: [release](https://mlr3spatial.mlr-org.com/) | [dev](https://mlr3spatial.mlr-org.com/dev/)
24 |
25 |
26 | [](https://github.com/mlr-org/mlr3spatial/actions/workflows/r-cmd-check.yml)
27 | [](https://CRAN.R-project.org/package=mlr3spatial)
28 | [](https://stackoverflow.com/questions/tagged/mlr3)
29 | [](https://lmmisld-lmu-stats-slds.srv.mwn.de/mlr_invite/)
30 |
31 |
32 | *mlr3spatial* is the package for spatial objects within the [`mlr3`](https://mlr-org.com) ecosystem.
33 | The package directly loads data from [`sf`](https://CRAN.R-project.org/package=sf) objects to train any mlr3 learner.
34 | The learner can predict on various raster formats ([`terra`](https://CRAN.R-project.org/package=terra), [`raster`](https://CRAN.R-project.org/package=raster) and [`stars`](https://CRAN.R-project.org/package=stars)) and writes the prediction raster to disk.
35 | mlr3spatial reads large raster objects in chunks to avoid memory issues and predicts the chunks in parallel.
36 | Check out [`mlr3spatiotempcv`](https://github.com/mlr-org/mlr3spatiotempcv) for spatiotemporal resampling within mlr3.
37 |
38 | ## Resources
39 |
40 | There are sections about spatial data in the [mlr3book](https://mlr3book.mlr-org.com).
41 |
42 | * Learn how to [predict](https://mlr3book.mlr-org.com/chapters/chapter13/beyond_regression_and_classification.html#sec-spatial-prediction) a spatial raster image.
43 | * Estimate the performance of a model with [spatial cross-validation](https://mlr3book.mlr-org.com/chapters/chapter13/beyond_regression_and_classification.html#sec-spatiotemporal).
44 |
45 | The gallery features articles about spatial data in the mlr3 ecosystem.
46 |
47 | * Learn the basics with a [land cover classification](https://mlr-org.com/gallery/technical/2023-02-27-land-cover-classification/) of the city of Leipzig.
48 |
49 | ## Installation
50 |
51 | Install the last release from CRAN:
52 |
53 | ```{r, eval=FALSE}
54 | install.packages("mlr3spatial")
55 | ```
56 |
57 | Install the development version from GitHub:
58 |
59 | ```{r, eval=FALSE}
60 | remotes::install_github("mlr-org/mlr3spatial")
61 | ```
62 |
63 | ## Example
64 |
65 | Our goal is to map the land cover of the city of Leipzig.
66 | The `mlr3spatial` package contains a Sentinel-2 scene of the city of Leipzig and a point vector with training sites.
67 | The Sentinel-2 scene is a 10m resolution multispectral image with 7 bands and the NDVI.
68 | The points represent samples of the four land cover classes: Forest, Pastures, Urban and Water.
69 | We load the raster with the [`terra`](https://CRAN.R-project.org/package=terra) package and the vector with the [`sf`](https://CRAN.R-project.org/package=sf) package in the R Session.
70 |
71 | ```{r}
72 | library(mlr3verse)
73 | library(mlr3spatial)
74 | library(terra, exclude = "resample")
75 | library(sf)
76 |
77 | leipzig = read_sf(system.file("extdata", "leipzig_points.gpkg", package = "mlr3spatial"), stringsAsFactors = TRUE)
78 |
79 | leipzig_raster = rast(system.file("extdata", "leipzig_raster.tif", package = "mlr3spatial"))
80 | ```
81 |
82 | The function `as_task_classif_st()` converts the `sf::sf` object to a spatial classification task.
83 |
84 | ```{r}
85 | task = as_task_classif_st(leipzig, target = "land_cover")
86 | task
87 | ```
88 |
89 | The points are located in the district of Lindenau and Zentrum-West.
90 |
91 | 
112 |
113 | Now we train a classification tree on the leipzig task.
114 |
115 | ``` r
116 | learner = lrn("classif.rpart")
117 | learner$train(task)
118 | ```
119 |
120 | As a last step, we predict the land cover class for the whole area of
121 | interest. For this, we pass the Sentinel-2 scene and the trained learner
122 | to the `predict_spatial()` function.
123 |
124 | ``` r
125 | land_cover = predict_spatial(leipzig_raster, learner)
126 | ```
127 |
128 | 
129 |
130 | ## FAQ
131 |
132 |
133 |
Eventually. It is not yet clear whether these would live in 137 | mlr3extralearners or in mlr3spatial. So far there are none yet. 138 |
139 | 134 | Will mlr3spatial support spatial learners? 135 |
136 |Eventually. It is not yet clear whether these would live in 137 | mlr3extralearners or in mlr3spatial. So far there are none yet. 138 |
140 |
mlr3spatiotempcv is solely devoted to resampling techniques. There 144 | are quite a few and keeping packages small is one of the development 145 | philosophies of the mlr3 framework. Also back in the days when 146 | mlr3spatiotempcv was developed, it was not yet clear how we want to 147 | structure additional spatial components such as prediction support for 148 | spatial classes and so on. 149 |
150 |
--------------------------------------------------------------------------------
/attic/benchmark.R:
--------------------------------------------------------------------------------
1 | library(mlr3spatial)
2 | library(mlr3learners)
3 | library(tictoc)
4 |
5 | # 10 MB - in memory ------------------------------------------------------------
6 |
7 | # train task
8 | stack = generate_stack(list(
9 | numeric_layer("x_1"),
10 | factor_layer("y", levels = c("a", "b"))),
11 | layer_size = 10)
12 | vector = sample_stack(stack, n = 100)
13 | task_train = as_task_classif_st(vector, id = "test_vector", target = "y")
14 | learner = lrn("classif.ranger")
15 | learner$train(task_train)
16 |
17 | # predict task
18 | dt = as.data.table(terra::values(stack))
19 | dt[, y := factor(y, levels = c("a", "b"))]
20 | backend = as_data_backend(dt)
21 | task = as_task_classif(dt, id = "test", target = "y")
22 |
23 | tic()
24 | pred = learner$predict(task) # 74 seconds
25 | toc()
26 |
27 |
28 | # 10 MB - terra ----------------------------------------------------------------
29 |
30 | # train task
31 | stack = generate_stack(list(
32 | numeric_layer("x_1"),
33 | factor_layer("y", levels = c("a", "b"))),
34 | layer_size = 10)
35 | vector = sample_stack(stack, n = 100)
36 | task_train = as_task_classif_st(vector, id = "test_vector", target = "y")
37 | learner = lrn("classif.ranger")
38 | learner$train(task_train)
39 |
40 | # predict task
41 | stack$y = NULL
42 | model = learner$model
43 |
44 | predfun = function(model, data) {
45 | library(ranger)
46 | predict(model, data)$predictions
47 | }
48 |
49 | tic()
50 | terra::predict(stack, model, fun = predfun) # 78 seconds
51 | toc()
52 |
53 | tic()
54 | terra::predict(stack, model, fun = predfun, cores = 4) # 36 seconds
55 | toc()
56 |
57 | tic()
58 | terra::predict(stack, learner, fun = predfun, cores = 8) # 47 seconds
59 | toc()
60 |
61 |
62 | # 10 MB - 1 chunk --------------------------------------------------------------
63 |
64 | # train task
65 | stack = generate_stack(list(
66 | numeric_layer("x_1"),
67 | factor_layer("y", levels = c("a", "b"))),
68 | layer_size = 10)
69 | vector = sample_stack(stack, n = 100)
70 | task_train = as_task_classif_st(vector, id = "test_vector", target = "y")
71 | learner = lrn("classif.ranger")
72 | learner$train(task_train)
73 |
74 | # predict task
75 | stack$y = NULL
76 | task_predict = as_task_unsupervised(stack, id = "test")
77 | learner$parallel_predict = TRUE
78 |
79 | ras = predict_spatial(task_predict, learner, chunksize = 10L) # 71 seconds
80 |
81 | with_future("multicore", workers = 2, {
82 | ras = predict_spatial(task_predict, learner, chunksize = 10L) # 38 seconds
83 | })
84 |
85 | with_future("multicore", workers = 4, {
86 | ras = predict_spatial(task_predict, learner, chunksize = 10L) # 21 seconds
87 | })
88 |
89 | with_future("multicore", workers = 8, {
90 | ras = predict_spatial(task_predict, learner, chunksize = 10L) # 19 seconds
91 | })
92 |
93 | # 50 MB - 5 chunks -------------------------------------------------------------
94 |
95 | # train task
96 | stack = generate_stack(list(
97 | numeric_layer("x_1"),
98 | factor_layer("y", levels = c("a", "b"))),
99 | layer_size = 50)
100 | vector = sample_stack(stack, n = 100)
101 | task_train = as_task_classif_st(vector, id = "test_vector", target = "y")
102 | learner = lrn("classif.ranger")
103 | learner$train(task_train)
104 |
105 | # predict task
106 | stack$y = NULL
107 | task_predict = as_task_unsupervised(stack, id = "test")
108 | learner$parallel_predict = TRUE
109 |
110 | ras = predict_spatial(task_predict, learner, chunksize = 10L) # 386 seconds
111 |
112 | with_future("multicore", workers = 2, {
113 | ras = predict_spatial(task_predict, learner, chunksize = 10L) #
114 | })
115 |
116 | with_future("multicore", workers = 4, {
117 | ras = predict_spatial(task_predict, learner, chunksize = 10L) # 142 seconds
118 | })
119 |
120 | with_future("multicore", workers = 8, {
121 | ras = predict_spatial(task_predict, learner, chunksize = 10L) #
122 | })
123 |
124 | # 20 MB - 1 chunk --------------------------------------------------------------
125 |
126 | # train task
127 | stack = generate_stack(list(
128 | numeric_layer("x_1"),
129 | factor_layer("y", levels = c("a", "b"))),
130 | layer_size = 20)
131 | vector = sample_stack(stack, n = 100)
132 | task_train = as_task_classif_st(vector, id = "test_vector", target = "y")
133 | learner = lrn("classif.ranger")
134 | learner$train(task_train)
135 |
136 | # predict task
137 | stack$y = NULL
138 | task_predict = as_task_unsupervised(stack, id = "test")
139 | learner$parallel_predict = TRUE
140 |
141 | ras = predict_spatial(task_predict, learner, chunksize = 20L) # 221 seconds
142 |
143 | with_future("multicore", workers = 2, {
144 | ras = predict_spatial(task_predict, learner, chunksize = 10L) #
145 | })
146 |
147 | with_future("multicore", workers = 4, {
148 | ras = predict_spatial(task_predict, learner, chunksize = 20L) # 61 seconds
149 | })
150 |
151 | with_future("multicore", workers = 8, {
152 | ras = predict_spatial(task_predict, learner, chunksize = 10L) #
153 | })
154 |
--------------------------------------------------------------------------------
/attic/benchmark.Rmd:
--------------------------------------------------------------------------------
1 | ---
2 | title: "Benchmarking parallel predictions"
3 | output: rmarkdown::html_vignette
4 | vignette: >
5 | %\VignetteIndexEntry{Benchmarking parallel predictions}
6 | %\VignetteEngine{knitr::rmarkdown}
7 | %\VignetteEncoding{UTF-8}
8 | ---
9 |
10 |
11 |
12 | This benchmark was run on a MacBook Pro 2021 (arm64) with the following specs
13 |
14 | - M1 Pro
15 | - 32 GB RAM
16 |
17 | Note that the differences between the parallel and sequential timings will increase for larger objects as the overhead for starting the parallel workers and collecting the results will decrease.
18 |
19 | It is not fully clear why the parallel approach of the {terra} package is slow than its sequential counterpart but it might relate to the single-core performance of the machine the benchmark was run on in combination with the overhead associated with starting the parallel cluster the way it is done in the {terra} package.
20 |
21 |
22 | ```r
23 | library(mlr3)
24 | library(mlr3spatial)
25 | library(future)
26 | library(bench)
27 | library(stars)
28 | #> Loading required package: abind
29 | #> Loading required package: sf
30 | #> Linking to GEOS 3.9.1, GDAL 3.4.2, PROJ 8.2.1; sf_use_s2() is TRUE
31 | library(rpart)
32 | ```
33 |
34 | ## Small files
35 |
36 | - ~ 250k values
37 | - 48 MB on disk
38 |
39 |
40 | ```r
41 | # SpatRaster demo stack
42 | stack_terra = demo_stack_spatraster(50)
43 | value = data.table::data.table(ID = c(0, 1), y = c("negative", "positive"))
44 | terra::set.cats(stack_terra, layer = "y", value = value)
45 | colnames = names(stack_terra)
46 | file_terra = tempfile("terra", fileext = ".tif")
47 | terra::writeRaster(stack_terra, file_terra)
48 |
49 | # RasterBrick demo stack
50 | stack_raster = demo_stack_rasterbrick(50)
51 | colnames_raster = names(stack_raster)
52 | file_raster = tempfile("raster", fileext = ".tif")
53 | raster::writeRaster(stack_raster, file_raster)
54 | #> Warning in .gd_SetProject(object, ...): NOT UPDATED FOR PROJ >= 6
55 |
56 | # tasks
57 | stack_terra = terra::rast(file_terra)
58 | backend_terra = DataBackendRaster$new(stack_terra)
59 | task_terra = as_task_regr(backend_terra, target = "x_1")
60 |
61 | stack_raster = raster::brick(file_raster)
62 | names(stack_raster) = colnames_raster
63 | backend_raster = DataBackendRaster$new(stack_raster)
64 | #> Warning in raster::wkt(from): no wkt comment
65 | task_raster = as_task_regr(backend_raster, target = "x_1")
66 |
67 | # Train learners
68 | set.seed(42)
69 | row_ids = sample(1:task_terra$nrow, 50)
70 |
71 | learner_task_terra = lrn("regr.rpart")
72 | learner_task_terra$parallel_predict = TRUE
73 | learner_task_terra$train(task_terra, row_ids = row_ids)
74 |
75 | learner_task_raster = lrn("regr.rpart")
76 | learner_task_terra$parallel_predict = TRUE
77 | learner_task_raster$train(task_raster, row_ids = row_ids)
78 |
79 | # non-mlr3 models
80 | rpart_task_terra = rpart::rpart(x_1 ~ ., task_terra$data(rows = row_ids))
81 | rpart_task_raster = rpart::rpart(x_1 ~ ., task_raster$data(rows = row_ids))
82 | ```
83 |
84 |
85 | ```r
86 | bm = bench::mark(
87 |
88 | "01-mlr3-terra-4-cores" = {
89 | plan(multicore, workers = 4)
90 | predict_spatial(task_terra, learner_task_terra, chunksize = 2000L)
91 | },
92 |
93 | "02-terra-4-cores" = terra::predict(stack_terra, rpart_task_terra, cores = 4, cpkgs = "rpart"),
94 |
95 | "03-mlr3-raster-4-cores" = {
96 | plan(multicore, workers = 4)
97 | predict_spatial(task_raster, learner_task_raster, chunksize = 2000L, format = "raster")
98 | },
99 |
100 | "04-raster-4-cores" = {
101 | library(raster)
102 | library(rpart)
103 | beginCluster(4, type = "PSOCK")
104 | clusterR(stack_raster, predict, args = list(model = rpart_task_raster))
105 | },
106 |
107 | check = FALSE, filter_gc = FALSE, min_iterations = 3,
108 | max_iterations = 3, memory = FALSE)
109 | #> Loading required package: sp
110 |
111 | bm$`itr/sec` = NULL
112 | bm$result = NULL
113 | bm$`gc/sec` = NULL
114 | bm$memory = NULL
115 | bm$mem_alloc = NULL
116 |
117 | print(bm)
118 | #> # A tibble: 4 × 8
119 | #> expression min median n_itr n_gc total_time time gc
120 | #> 141 | Why are there two packages, mlr3spatial and mlr3spatiotempcv? 142 |
143 |mlr3spatiotempcv is solely devoted to resampling techniques. There 144 | are quite a few and keeping packages small is one of the development 145 | philosophies of the mlr3 framework. Also back in the days when 146 | mlr3spatiotempcv was developed, it was not yet clear how we want to 147 | structure additional spatial components such as prediction support for 148 | spatial classes and so on. 149 |
mlr3::DataBackend$format()
61 | mlr3::DataBackend$print()
62 | mlr3::Task$add_strata()
45 | mlr3::Task$cbind()
46 | mlr3::Task$data()
47 | mlr3::Task$filter()
48 | mlr3::Task$format()
49 | mlr3::Task$formula()
50 | mlr3::Task$head()
51 | mlr3::Task$help()
52 | mlr3::Task$levels()
53 | mlr3::Task$missings()
54 | mlr3::Task$rbind()
55 | mlr3::Task$rename()
56 | mlr3::Task$select()
57 | mlr3::Task$set_col_roles()
58 | mlr3::Task$set_levels()
59 | mlr3::Task$set_row_roles()
60 | mlr3::TaskClassif$droplevels()
61 | mlr3::TaskClassif$truth()
62 | mlr3::Task$add_strata()
45 | mlr3::Task$cbind()
46 | mlr3::Task$data()
47 | mlr3::Task$droplevels()
48 | mlr3::Task$filter()
49 | mlr3::Task$format()
50 | mlr3::Task$formula()
51 | mlr3::Task$head()
52 | mlr3::Task$help()
53 | mlr3::Task$levels()
54 | mlr3::Task$missings()
55 | mlr3::Task$rbind()
56 | mlr3::Task$rename()
57 | mlr3::Task$select()
58 | mlr3::Task$set_col_roles()
59 | mlr3::Task$set_levels()
60 | mlr3::Task$set_row_roles()
61 | mlr3::TaskRegr$truth()
62 |
-
121 | #> 1 01-mlr3-terra-4-cores 852.65ms 915.91ms 3 22 2.77s
-
226 | #> 1 01-mlr3-terra-4-cores 11.93s 12.01s 3 167 37.23s
}}
28 | \describe{
29 | \item{\code{rownames}}{(\code{integer()})\cr
30 | Returns vector of all distinct row identifiers, i.e. the contents of the primary key column.}
31 |
32 | \item{\code{colnames}}{(\code{character()})\cr
33 | Returns vector of all column names.}
34 |
35 | \item{\code{nrow}}{(\code{integer(1)})\cr
36 | Number of rows (observations).}
37 |
38 | \item{\code{ncol}}{(\code{integer(1)})\cr
39 | Number of columns (variables).}
40 |
41 | \item{\code{stack}}{(\code{SpatRaster})\cr
42 | Raster stack.}
43 | }
44 | \if{html}{\out{
}}
45 | }
46 | \section{Methods}{
47 | \subsection{Public methods}{
48 | \itemize{
49 | \item \href{#method-DataBackendRaster-new}{\code{DataBackendRaster$new()}}
50 | \item \href{#method-DataBackendRaster-data}{\code{DataBackendRaster$data()}}
51 | \item \href{#method-DataBackendRaster-head}{\code{DataBackendRaster$head()}}
52 | \item \href{#method-DataBackendRaster-distinct}{\code{DataBackendRaster$distinct()}}
53 | \item \href{#method-DataBackendRaster-missings}{\code{DataBackendRaster$missings()}}
54 | \item \href{#method-DataBackendRaster-coordinates}{\code{DataBackendRaster$coordinates()}}
55 | }
56 | }
57 | \if{html}{\out{
58 | Inherited methods
59 |-
60 |
}} 66 | \if{html}{\out{}} 67 | \if{latex}{\out{\hypertarget{method-DataBackendRaster-new}{}}} 68 | \subsection{Method \code{new()}}{ 69 | Creates a new instance of this \link[R6:R6Class]{R6} class. 70 | \subsection{Usage}{ 71 | \if{html}{\out{
}}\preformatted{DataBackendRaster$new(data)}\if{html}{\out{
}}
72 | }
73 |
74 | \subsection{Arguments}{
75 | \if{html}{\out{}}
76 | \describe{
77 | \item{\code{data}}{(\link[terra:SpatRaster-class]{terra::SpatRaster})\cr
78 | The input \link[terra:SpatRaster-class]{terra::SpatRaster}.}
79 | }
80 | \if{html}{\out{
}}
81 | }
82 | }
83 | \if{html}{\out{}} 84 | \if{html}{\out{}} 85 | \if{latex}{\out{\hypertarget{method-DataBackendRaster-data}{}}} 86 | \subsection{Method \code{data()}}{ 87 | Returns a slice of the raster in the specified format. 88 | Currently, the only supported formats is \code{"data.table"}. 89 | 90 | The rows must be addressed as vector of cells indices, columns must be 91 | referred to via layer names. Queries for rows with no matching row id and 92 | queries for columns with no matching column name are silently ignored. 93 | 94 | Rows are guaranteed to be returned in the same order as \code{rows}, columns 95 | may be returned in an arbitrary order. Duplicated row ids result in 96 | duplicated rows, duplicated column names lead to an exception. 97 | \subsection{Usage}{ 98 | \if{html}{\out{
}}\preformatted{DataBackendRaster$data(rows, cols, data_format = "data.table")}\if{html}{\out{
}}
99 | }
100 |
101 | \subsection{Arguments}{
102 | \if{html}{\out{}}
103 | \describe{
104 | \item{\code{rows}}{\code{integer()}\cr
105 | Row indices. Row indices start with 1 in the upper left corner in the
106 | raster, increase from left to right and then from top to bottom. The last
107 | cell is in the bottom right corner and the row index equals the number of
108 | cells in the raster.}
109 |
110 | \item{\code{cols}}{\code{character()}\cr
111 | Column names.}
112 |
113 | \item{\code{data_format}}{(\code{character(1)})\cr
114 | Desired data format. Currently only \code{"data.table"} supported.}
115 | }
116 | \if{html}{\out{
}}
117 | }
118 | }
119 | \if{html}{\out{}} 120 | \if{html}{\out{}} 121 | \if{latex}{\out{\hypertarget{method-DataBackendRaster-head}{}}} 122 | \subsection{Method \code{head()}}{ 123 | Retrieve the first \code{n} rows. 124 | \subsection{Usage}{ 125 | \if{html}{\out{
}}\preformatted{DataBackendRaster$head(n = 6L)}\if{html}{\out{
}}
126 | }
127 |
128 | \subsection{Arguments}{
129 | \if{html}{\out{}}
130 | \describe{
131 | \item{\code{n}}{(\code{integer(1)})\cr
132 | Number of rows.}
133 | }
134 | \if{html}{\out{
}}
135 | }
136 | \subsection{Returns}{
137 | \code{\link[data.table:data.table]{data.table::data.table()}} of the first \code{n} rows.
138 | }
139 | }
140 | \if{html}{\out{}} 141 | \if{html}{\out{}} 142 | \if{latex}{\out{\hypertarget{method-DataBackendRaster-distinct}{}}} 143 | \subsection{Method \code{distinct()}}{ 144 | Returns a named list of vectors of distinct values for each column 145 | specified. If \code{na_rm} is \code{TRUE}, missing values are removed from the 146 | returned vectors of distinct values. Non-existing rows and columns are 147 | silently ignored. 148 | \subsection{Usage}{ 149 | \if{html}{\out{
}}\preformatted{DataBackendRaster$distinct(rows, cols, na_rm = TRUE)}\if{html}{\out{
}}
150 | }
151 |
152 | \subsection{Arguments}{
153 | \if{html}{\out{}}
154 | \describe{
155 | \item{\code{rows}}{\code{integer()}\cr
156 | Row indices. Row indices start with 1 in the upper left corner in the
157 | raster, increase from left to right and then from top to bottom. The last
158 | cell is in the bottom right corner and the row index equals the number of
159 | cells in the raster.}
160 |
161 | \item{\code{cols}}{\code{character()}\cr
162 | Column names.}
163 |
164 | \item{\code{na_rm}}{\code{logical(1)}\cr
165 | Whether to remove NAs or not.}
166 | }
167 | \if{html}{\out{
}}
168 | }
169 | \subsection{Returns}{
170 | Named \code{list()} of distinct values.
171 | }
172 | }
173 | \if{html}{\out{}} 174 | \if{html}{\out{}} 175 | \if{latex}{\out{\hypertarget{method-DataBackendRaster-missings}{}}} 176 | \subsection{Method \code{missings()}}{ 177 | Returns the number of missing values per column in the specified slice 178 | of data. Non-existing rows and columns are silently ignored. 179 | \subsection{Usage}{ 180 | \if{html}{\out{
}}\preformatted{DataBackendRaster$missings(rows, cols)}\if{html}{\out{
}}
181 | }
182 |
183 | \subsection{Arguments}{
184 | \if{html}{\out{}}
185 | \describe{
186 | \item{\code{rows}}{\code{integer()}\cr
187 | Row indices. Row indices start with 1 in the upper left corner in the
188 | raster, increase from left to right and then from top to bottom. The last
189 | cell is in the bottom right corner and the row index equals the number of
190 | cells in the raster.}
191 |
192 | \item{\code{cols}}{\code{character()}\cr
193 | Column names.}
194 | }
195 | \if{html}{\out{
}}
196 | }
197 | \subsection{Returns}{
198 | Total of missing values per column (named \code{numeric()}).
199 | }
200 | }
201 | \if{html}{\out{}} 202 | \if{html}{\out{}} 203 | \if{latex}{\out{\hypertarget{method-DataBackendRaster-coordinates}{}}} 204 | \subsection{Method \code{coordinates()}}{ 205 | Returns the coordinates of \code{rows}. 206 | If \code{rows} is missing, all coordinates are returned. 207 | \subsection{Usage}{ 208 | \if{html}{\out{
}}\preformatted{DataBackendRaster$coordinates(rows)}\if{html}{\out{
}}
209 | }
210 |
211 | \subsection{Arguments}{
212 | \if{html}{\out{}}
213 | \describe{
214 | \item{\code{rows}}{\code{integer()}\cr
215 | Row indices. Row indices start with 1 in the upper left corner in the
216 | raster, increase from left to right and then from top to bottom. The last
217 | cell is in the bottom right corner and the row index equals the number of
218 | cells in the raster.}
219 | }
220 | \if{html}{\out{
}}
221 | }
222 | \subsection{Returns}{
223 | \code{\link[data.table:data.table]{data.table::data.table()}} of coordinates of \code{rows}.
224 | }
225 | }
226 | }
227 |
--------------------------------------------------------------------------------
/man/DataBackendVector.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/DataBackendVector.R
3 | \name{DataBackendVector}
4 | \alias{DataBackendVector}
5 | \title{DataBackend for Vector Objects}
6 | \description{
7 | \link[mlr3:DataBackend]{mlr3::DataBackend} for \link[sf:sf]{sf::sf} vector objects.
8 | }
9 | \section{Super classes}{
10 | \code{\link[mlr3:DataBackend]{mlr3::DataBackend}} -> \code{\link[mlr3:DataBackendDataTable]{mlr3::DataBackendDataTable}} -> \code{DataBackendVector}
11 | }
12 | \section{Active bindings}{
13 | \if{html}{\out{}}
14 | \describe{
15 | \item{\code{sfc}}{(\link[sf:sfc]{sf::sfc})\cr
16 | Returns the sfc object.}
17 | }
18 | \if{html}{\out{
}}
19 | }
20 | \section{Methods}{
21 | \subsection{Public methods}{
22 | \itemize{
23 | \item \href{#method-DataBackendVector-new}{\code{DataBackendVector$new()}}
24 | }
25 | }
26 | \if{html}{\out{
27 | Inherited methods
28 | 36 |}} 39 | \if{html}{\out{}} 40 | \if{latex}{\out{\hypertarget{method-DataBackendVector-new}{}}} 41 | \subsection{Method \code{new()}}{ 42 | Creates a new instance of this \link[R6:R6Class]{R6} class. 43 | \subsection{Usage}{ 44 | \if{html}{\out{
}}\preformatted{DataBackendVector$new(data, primary_key)}\if{html}{\out{
}}
45 | }
46 |
47 | \subsection{Arguments}{
48 | \if{html}{\out{}}
49 | \describe{
50 | \item{\code{data}}{(\code{sf})\cr
51 | A raster object.}
52 |
53 | \item{\code{primary_key}}{(\code{character(1)} | \code{integer()})\cr
54 | Name of the primary key column, or integer vector of row ids.}
55 | }
56 | \if{html}{\out{
}}
57 | }
58 | }
59 | }
60 |
--------------------------------------------------------------------------------
/man/TaskClassifST.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/TaskClassifST.R
3 | \name{TaskClassifST}
4 | \alias{TaskClassifST}
5 | \title{Spatiotemporal Classification Task}
6 | \description{
7 | This task specializes \link{TaskClassif} for spatiotemporal classification problems.
8 |
9 | A spatial example task is available via \code{tsk("ecuador")}.
10 |
11 | The coordinate reference system passed during initialization must match the one which was used during data creation, otherwise offsets of multiple meters may occur.
12 | By default, coordinates are not used as features.
13 | This can be changed by setting \code{coords_as_features = TRUE}.
14 | }
15 | \section{Super classes}{
16 | \code{\link[mlr3:Task]{mlr3::Task}} -> \code{\link[mlr3:TaskSupervised]{mlr3::TaskSupervised}} -> \code{\link[mlr3:TaskClassif]{mlr3::TaskClassif}} -> \code{TaskClassifST}
17 | }
18 | \section{Active bindings}{
19 | \if{html}{\out{}}
20 | \describe{
21 | \item{\code{crs}}{(\code{character(1)})\cr
22 | Returns coordinate reference system of task.}
23 |
24 | \item{\code{coordinate_names}}{(\code{character()})\cr
25 | Returns coordinate names.}
26 |
27 | \item{\code{coords_as_features}}{(\code{logical(1)})\cr
28 | If \code{TRUE}, coordinates are used as features.}
29 | }
30 | \if{html}{\out{
}}
31 | }
32 | \section{Methods}{
33 | \subsection{Public methods}{
34 | \itemize{
35 | \item \href{#method-TaskClassifST-new}{\code{TaskClassifST$new()}}
36 | \item \href{#method-TaskClassifST-coordinates}{\code{TaskClassifST$coordinates()}}
37 | \item \href{#method-TaskClassifST-print}{\code{TaskClassifST$print()}}
38 | \item \href{#method-TaskClassifST-clone}{\code{TaskClassifST$clone()}}
39 | }
40 | }
41 | \if{html}{\out{
42 | Inherited methods
43 |-
44 |
}} 66 | \if{html}{\out{}} 67 | \if{latex}{\out{\hypertarget{method-TaskClassifST-new}{}}} 68 | \subsection{Method \code{new()}}{ 69 | Creates a new instance of this \link[R6:R6Class]{R6} class. 70 | The function \code{\link[=as_task_classif_st]{as_task_classif_st()}} provides an alternative way to construct classification tasks. 71 | \subsection{Usage}{ 72 | \if{html}{\out{
}}\preformatted{TaskClassifST$new(
73 | id,
74 | backend,
75 | target,
76 | positive = NULL,
77 | label = NA_character_,
78 | coordinate_names,
79 | crs = NA_character_,
80 | coords_as_features = FALSE,
81 | extra_args = list()
82 | )}\if{html}{\out{
}}
83 | }
84 |
85 | \subsection{Arguments}{
86 | \if{html}{\out{}}
87 | \describe{
88 | \item{\code{id}}{(\code{character(1)})\cr
89 | Identifier for the new instance.}
90 |
91 | \item{\code{backend}}{(\link{DataBackend})\cr
92 | Either a \link{DataBackend}, or any object which is convertible to a \link{DataBackend} with \code{as_data_backend()}.
93 | E.g., am \code{sf} will be converted to a \link{DataBackendDataTable}.}
94 |
95 | \item{\code{target}}{(\code{character(1)})\cr
96 | Name of the target column.}
97 |
98 | \item{\code{positive}}{(\code{character(1)})\cr
99 | Only for binary classification: Name of the positive class.
100 | The levels of the target columns are reordered accordingly, so that the first element of \verb{$class_names} is the positive class, and the second element is the negative class.}
101 |
102 | \item{\code{label}}{(\code{character(1)})\cr
103 | Label for the new instance.}
104 |
105 | \item{\code{coordinate_names}}{(\code{character(1)})\cr
106 | The column names of the coordinates in the data.}
107 |
108 | \item{\code{crs}}{(\code{character(1)})\cr
109 | Coordinate reference system.
110 | WKT2 or EPSG string.}
111 |
112 | \item{\code{coords_as_features}}{(\code{logical(1)})\cr
113 | If \code{TRUE}, coordinates are used as features.}
114 |
115 | \item{\code{extra_args}}{(named \code{list()})\cr
116 | Named list of constructor arguments, required for converting task types via \code{\link[=convert_task]{convert_task()}}.}
117 | }
118 | \if{html}{\out{
}}
119 | }
120 | }
121 | \if{html}{\out{}} 122 | \if{html}{\out{}} 123 | \if{latex}{\out{\hypertarget{method-TaskClassifST-coordinates}{}}} 124 | \subsection{Method \code{coordinates()}}{ 125 | Returns coordinates of observations. 126 | \subsection{Usage}{ 127 | \if{html}{\out{
}}\preformatted{TaskClassifST$coordinates(row_ids = NULL)}\if{html}{\out{
}}
128 | }
129 |
130 | \subsection{Arguments}{
131 | \if{html}{\out{}}
132 | \describe{
133 | \item{\code{row_ids}}{(\code{integer()})\cr
134 | Vector of rows indices as subset of \code{task$row_ids}.}
135 | }
136 | \if{html}{\out{
}}
137 | }
138 | \subsection{Returns}{
139 | \code{\link[data.table:data.table]{data.table::data.table()}}
140 | }
141 | }
142 | \if{html}{\out{}} 143 | \if{html}{\out{}} 144 | \if{latex}{\out{\hypertarget{method-TaskClassifST-print}{}}} 145 | \subsection{Method \code{print()}}{ 146 | Print the task. 147 | \subsection{Usage}{ 148 | \if{html}{\out{
}}\preformatted{TaskClassifST$print(...)}\if{html}{\out{
}}
149 | }
150 |
151 | \subsection{Arguments}{
152 | \if{html}{\out{}}
153 | \describe{
154 | \item{\code{...}}{Arguments passed to the \verb{$print()} method of the superclass.}
155 | }
156 | \if{html}{\out{
}}
157 | }
158 | }
159 | \if{html}{\out{}} 160 | \if{html}{\out{}} 161 | \if{latex}{\out{\hypertarget{method-TaskClassifST-clone}{}}} 162 | \subsection{Method \code{clone()}}{ 163 | The objects of this class are cloneable with this method. 164 | \subsection{Usage}{ 165 | \if{html}{\out{
}}\preformatted{TaskClassifST$clone(deep = FALSE)}\if{html}{\out{
}}
166 | }
167 |
168 | \subsection{Arguments}{
169 | \if{html}{\out{}}
170 | \describe{
171 | \item{\code{deep}}{Whether to make a deep clone.}
172 | }
173 | \if{html}{\out{
}}
174 | }
175 | }
176 | }
177 |
--------------------------------------------------------------------------------
/man/TaskRegrST.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/TaskRegrST.R
3 | \name{TaskRegrST}
4 | \alias{TaskRegrST}
5 | \title{Spatiotemporal Regression Task}
6 | \description{
7 | This task specializes \link{TaskRegr} for spatiotemporal regression problems.
8 |
9 | A spatial example task is available via \code{tsk("cookfarm_mlr3")}.
10 |
11 | The coordinate reference system passed during initialization must match the one which was used during data creation, otherwise offsets of multiple meters may occur.
12 | By default, coordinates are not used as features.
13 | This can be changed by setting \code{coords_as_features = TRUE}.
14 | }
15 | \section{Super classes}{
16 | \code{\link[mlr3:Task]{mlr3::Task}} -> \code{\link[mlr3:TaskSupervised]{mlr3::TaskSupervised}} -> \code{\link[mlr3:TaskRegr]{mlr3::TaskRegr}} -> \code{TaskRegrST}
17 | }
18 | \section{Active bindings}{
19 | \if{html}{\out{}}
20 | \describe{
21 | \item{\code{crs}}{(\code{character(1)})\cr
22 | Returns coordinate reference system of the task.}
23 |
24 | \item{\code{coordinate_names}}{(\code{character()})\cr
25 | Returns coordinate names.}
26 |
27 | \item{\code{coords_as_features}}{(\code{logical(1)})\cr
28 | If \code{TRUE}, coordinates are used as features.}
29 | }
30 | \if{html}{\out{
}}
31 | }
32 | \section{Methods}{
33 | \subsection{Public methods}{
34 | \itemize{
35 | \item \href{#method-TaskRegrST-new}{\code{TaskRegrST$new()}}
36 | \item \href{#method-TaskRegrST-coordinates}{\code{TaskRegrST$coordinates()}}
37 | \item \href{#method-TaskRegrST-print}{\code{TaskRegrST$print()}}
38 | \item \href{#method-TaskRegrST-clone}{\code{TaskRegrST$clone()}}
39 | }
40 | }
41 | \if{html}{\out{
42 | Inherited methods
43 |-
44 |
}} 66 | \if{html}{\out{}} 67 | \if{latex}{\out{\hypertarget{method-TaskRegrST-new}{}}} 68 | \subsection{Method \code{new()}}{ 69 | Creates a new instance of this \link[R6:R6Class]{R6} class. 70 | The function \code{\link[=as_task_regr_st]{as_task_regr_st()}} provides an alternative way to construct classification tasks. 71 | \subsection{Usage}{ 72 | \if{html}{\out{
}}\preformatted{TaskRegrST$new(
73 | id,
74 | backend,
75 | target,
76 | label = NA_character_,
77 | coordinate_names,
78 | crs = NA_character_,
79 | coords_as_features = FALSE,
80 | extra_args = list()
81 | )}\if{html}{\out{
}}
82 | }
83 |
84 | \subsection{Arguments}{
85 | \if{html}{\out{}}
86 | \describe{
87 | \item{\code{id}}{(\code{character(1)})\cr
88 | Identifier for the new instance.}
89 |
90 | \item{\code{backend}}{(\link{DataBackend})\cr
91 | Either a \link{DataBackend}, or any object which is convertible to a \link{DataBackend} with \code{as_data_backend()}.
92 | E.g., am \code{sf} will be converted to a \link{DataBackendDataTable}.}
93 |
94 | \item{\code{target}}{(\code{character(1)})\cr
95 | Name of the target column.}
96 |
97 | \item{\code{label}}{(\code{character(1)})\cr
98 | Label for the new instance.}
99 |
100 | \item{\code{coordinate_names}}{(\code{character(1)})\cr
101 | The column names of the coordinates in the data.}
102 |
103 | \item{\code{crs}}{(\code{character(1)})\cr
104 | Coordinate reference system.
105 | WKT2 or EPSG string.}
106 |
107 | \item{\code{coords_as_features}}{(\code{logical(1)})\cr
108 | If \code{TRUE}, coordinates are used as features.}
109 |
110 | \item{\code{extra_args}}{(named \code{list()})\cr
111 | Named list of constructor arguments, required for converting task types via \code{\link[=convert_task]{convert_task()}}.}
112 | }
113 | \if{html}{\out{
}}
114 | }
115 | }
116 | \if{html}{\out{}} 117 | \if{html}{\out{}} 118 | \if{latex}{\out{\hypertarget{method-TaskRegrST-coordinates}{}}} 119 | \subsection{Method \code{coordinates()}}{ 120 | Returns coordinates of observations. 121 | \subsection{Usage}{ 122 | \if{html}{\out{
}}\preformatted{TaskRegrST$coordinates(row_ids = NULL)}\if{html}{\out{
}}
123 | }
124 |
125 | \subsection{Arguments}{
126 | \if{html}{\out{}}
127 | \describe{
128 | \item{\code{row_ids}}{(\code{integer()})\cr
129 | Vector of rows indices as subset of \code{task$row_ids}.}
130 | }
131 | \if{html}{\out{
}}
132 | }
133 | \subsection{Returns}{
134 | \code{\link[data.table:data.table]{data.table::data.table()}}
135 | }
136 | }
137 | \if{html}{\out{}} 138 | \if{html}{\out{}} 139 | \if{latex}{\out{\hypertarget{method-TaskRegrST-print}{}}} 140 | \subsection{Method \code{print()}}{ 141 | Print the task. 142 | \subsection{Usage}{ 143 | \if{html}{\out{
}}\preformatted{TaskRegrST$print(...)}\if{html}{\out{
}}
144 | }
145 |
146 | \subsection{Arguments}{
147 | \if{html}{\out{}}
148 | \describe{
149 | \item{\code{...}}{Arguments passed to the \verb{$print()} method of the superclass.}
150 | }
151 | \if{html}{\out{
}}
152 | }
153 | }
154 | \if{html}{\out{}} 155 | \if{html}{\out{}} 156 | \if{latex}{\out{\hypertarget{method-TaskRegrST-clone}{}}} 157 | \subsection{Method \code{clone()}}{ 158 | The objects of this class are cloneable with this method. 159 | \subsection{Usage}{ 160 | \if{html}{\out{
}}\preformatted{TaskRegrST$clone(deep = FALSE)}\if{html}{\out{
}}
161 | }
162 |
163 | \subsection{Arguments}{
164 | \if{html}{\out{}}
165 | \describe{
166 | \item{\code{deep}}{Whether to make a deep clone.}
167 | }
168 | \if{html}{\out{
}}
169 | }
170 | }
171 | }
172 |
--------------------------------------------------------------------------------
/man/as_data_backend.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/DataBackendRaster.R, R/DataBackendVector.R
3 | \name{as_data_backend.stars}
4 | \alias{as_data_backend.stars}
5 | \alias{as_data_backend.SpatRaster}
6 | \alias{as_data_backend.RasterBrick}
7 | \alias{as_data_backend.RasterStack}
8 | \alias{as_data_backend.sf}
9 | \title{Coerce to spatial DataBackend}
10 | \usage{
11 | \method{as_data_backend}{stars}(data, primary_key = NULL, ...)
12 |
13 | \method{as_data_backend}{SpatRaster}(data, primary_key = NULL, ...)
14 |
15 | \method{as_data_backend}{RasterBrick}(data, primary_key = NULL, ...)
16 |
17 | \method{as_data_backend}{RasterStack}(data, primary_key = NULL, ...)
18 |
19 | \method{as_data_backend}{sf}(data, primary_key = NULL, keep_rownames = FALSE, ...)
20 | }
21 | \arguments{
22 | \item{data}{(\link[terra:SpatRaster-class]{terra::SpatRaster})\cr
23 | The input \link[terra:SpatRaster-class]{terra::SpatRaster}.}
24 |
25 | \item{primary_key}{(\code{character(1)} | \code{integer()})\cr
26 | Name of the primary key column, or integer vector of row ids.}
27 |
28 | \item{...}{(\code{any})\cr
29 | Not used.}
30 |
31 | \item{keep_rownames}{(\code{logical(1)} | \code{character(1)})\cr
32 | If \code{TRUE} or a single string, keeps the row names of \code{data} as a new column.
33 | The column is named like the provided string, defaulting to \code{"..rownames"} for \code{keep_rownames == TRUE}.
34 | Note that the created column will be used as a regular feature by the task unless you manually change the column role.
35 | Also see \code{\link[data.table:as.data.table]{data.table::as.data.table()}}.}
36 | }
37 | \value{
38 | \link{DataBackend}.
39 | }
40 | \description{
41 | Wraps a \link{DataBackend} around spatial objects.
42 | Currently these S3 methods are only alternative ways for writing \code{DataBackendRaster$new()}.
43 | They do not support coercing from other backends yet.
44 | }
45 |
--------------------------------------------------------------------------------
/man/as_task_classif_st.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/as_task_classif_st.R
3 | \name{as_task_classif_st}
4 | \alias{as_task_classif_st}
5 | \alias{as_task_classif_st.TaskClassifST}
6 | \alias{as_task_classif_st.data.frame}
7 | \alias{as_task_classif_st.DataBackend}
8 | \alias{as_task_classif_st.sf}
9 | \alias{as_task_classif_st.TaskRegrST}
10 | \title{Convert to a Spatiotemporal Classification Task}
11 | \usage{
12 | as_task_classif_st(x, ...)
13 |
14 | \method{as_task_classif_st}{TaskClassifST}(x, clone = FALSE, ...)
15 |
16 | \method{as_task_classif_st}{data.frame}(
17 | x,
18 | target,
19 | id = deparse(substitute(x)),
20 | positive = NULL,
21 | coordinate_names,
22 | crs = NA_character_,
23 | coords_as_features = FALSE,
24 | label = NA_character_,
25 | ...
26 | )
27 |
28 | \method{as_task_classif_st}{DataBackend}(
29 | x,
30 | target,
31 | id = deparse(substitute(x)),
32 | positive = NULL,
33 | coordinate_names,
34 | crs,
35 | coords_as_features = FALSE,
36 | label = NA_character_,
37 | ...
38 | )
39 |
40 | \method{as_task_classif_st}{sf}(
41 | x,
42 | target = NULL,
43 | id = deparse(substitute(x)),
44 | positive = NULL,
45 | coords_as_features = FALSE,
46 | label = NA_character_,
47 | ...
48 | )
49 |
50 | \method{as_task_classif_st}{TaskRegrST}(
51 | x,
52 | target = NULL,
53 | drop_original_target = FALSE,
54 | drop_levels = TRUE,
55 | ...
56 | )
57 | }
58 | \arguments{
59 | \item{x}{(any)\cr
60 | Object to convert.}
61 |
62 | \item{...}{(any)\cr
63 | Additional arguments.}
64 |
65 | \item{clone}{(\code{logical(1)})\cr
66 | If \code{TRUE}, ensures that the returned object is not the same as the input \code{x}.}
67 |
68 | \item{target}{(\code{character(1)})\cr
69 | Name of the target column.}
70 |
71 | \item{id}{(\code{character(1)})\cr
72 | Id for the new task.
73 | Defaults to the (deparsed and substituted) name of the data argument.}
74 |
75 | \item{positive}{(\code{character(1)})\cr
76 | Level of the positive class. See \link[mlr3]{TaskClassif}.}
77 |
78 | \item{coordinate_names}{(\code{character(1)})\cr
79 | The column names of the coordinates in the data.}
80 |
81 | \item{crs}{(\code{character(1)})\cr
82 | Coordinate reference system.
83 | WKT2 or EPSG string.}
84 |
85 | \item{coords_as_features}{(\code{logical(1)})\cr
86 | If \code{TRUE}, coordinates are used as features.}
87 |
88 | \item{label}{(\code{character(1)})\cr
89 | Label for the new instance.}
90 |
91 | \item{drop_original_target}{(\code{logical(1)})\cr
92 | If \code{FALSE} (default), the original target is added as a feature.
93 | Otherwise the original target is dropped.}
94 |
95 | \item{drop_levels}{(\code{logical(1)})\cr
96 | If \code{TRUE} (default), unused levels of the new target variable are dropped.}
97 | }
98 | \value{
99 | \link{TaskClassifST}
100 | }
101 | \description{
102 | Convert object to a \link{TaskClassifST}.
103 | This is a S3 generic, specialized for at least the following objects:
104 | \enumerate{
105 | \item \link{TaskClassifST}: Ensure the identity.
106 | \item \code{\link[=data.frame]{data.frame()}} and \link{DataBackend}: Provides an alternative to the constructor of \link{TaskClassifST}.
107 | \item \link[sf:sf]{sf::sf}: Extracts spatial meta data before construction.
108 | \item \link{TaskRegr}: Calls \code{\link[=convert_task]{convert_task()}}.
109 | }
110 | }
111 |
--------------------------------------------------------------------------------
/man/as_task_regr_st.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/as_task_regr_st.R
3 | \name{as_task_regr_st}
4 | \alias{as_task_regr_st}
5 | \alias{as_task_regr_st.TaskRegrST}
6 | \alias{as_task_regr_st.data.frame}
7 | \alias{as_task_regr_st.DataBackend}
8 | \alias{as_task_regr_st.sf}
9 | \alias{as_task_regr_st.TaskClassifST}
10 | \title{Convert to a Spatiotemporal Regression Task}
11 | \usage{
12 | as_task_regr_st(x, ...)
13 |
14 | \method{as_task_regr_st}{TaskRegrST}(x, clone = FALSE, ...)
15 |
16 | \method{as_task_regr_st}{data.frame}(
17 | x,
18 | target,
19 | id = deparse(substitute(x)),
20 | coordinate_names,
21 | crs = NA_character_,
22 | coords_as_features = FALSE,
23 | label = NA_character_,
24 | ...
25 | )
26 |
27 | \method{as_task_regr_st}{DataBackend}(
28 | x,
29 | target,
30 | id = deparse(substitute(x)),
31 | coordinate_names,
32 | crs,
33 | coords_as_features = FALSE,
34 | label = NA_character_,
35 | ...
36 | )
37 |
38 | \method{as_task_regr_st}{sf}(
39 | x,
40 | target = NULL,
41 | id = deparse(substitute(x)),
42 | coords_as_features = FALSE,
43 | label = NA_character_,
44 | ...
45 | )
46 |
47 | \method{as_task_regr_st}{TaskClassifST}(
48 | x,
49 | target = NULL,
50 | drop_original_target = FALSE,
51 | drop_levels = TRUE,
52 | ...
53 | )
54 | }
55 | \arguments{
56 | \item{x}{(any)\cr
57 | Object to convert.}
58 |
59 | \item{...}{(any)\cr
60 | Additional arguments.}
61 |
62 | \item{clone}{(\code{logical(1)})\cr
63 | If \code{TRUE}, ensures that the returned object is not the same as the input \code{x}.}
64 |
65 | \item{target}{(\code{character(1)})\cr
66 | Name of the target column.}
67 |
68 | \item{id}{(\code{character(1)})\cr
69 | Id for the new task.
70 | Defaults to the (deparsed and substituted) name of the data argument.}
71 |
72 | \item{coordinate_names}{(\code{character(1)})\cr
73 | The column names of the coordinates in the data.}
74 |
75 | \item{crs}{(\code{character(1)})\cr
76 | Coordinate reference system.
77 | WKT2 or EPSG string.}
78 |
79 | \item{coords_as_features}{(\code{logical(1)})\cr
80 | If \code{TRUE}, coordinates are used as features.}
81 |
82 | \item{label}{(\code{character(1)})\cr
83 | Label for the new instance.}
84 |
85 | \item{drop_original_target}{(\code{logical(1)})\cr
86 | If \code{FALSE} (default), the original target is added as a feature.
87 | Otherwise the original target is dropped.}
88 |
89 | \item{drop_levels}{(\code{logical(1)})\cr
90 | If \code{TRUE} (default), unused levels of the new target variable are dropped.}
91 | }
92 | \value{
93 | \link{TaskRegrST}
94 | }
95 | \description{
96 | Convert object to a \link{TaskRegrST}.
97 | This is a S3 generic, specialized for at least the following objects:
98 | \enumerate{
99 | \item \link{TaskRegrST}: Ensure the identity.
100 | \item \code{\link[=data.frame]{data.frame()}} and \link{DataBackend}: Provides an alternative to the constructor of \link{TaskRegrST}.
101 | \item \link[sf:sf]{sf::sf}: Extracts spatial meta data before construction.
102 | \item \link{TaskClassif}: Calls \code{\link[=convert_task]{convert_task()}}.
103 | }
104 | }
105 |
--------------------------------------------------------------------------------
/man/block_size.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/helper.R
3 | \name{block_size}
4 | \alias{block_size}
5 | \title{Split Raster Into Chunks}
6 | \usage{
7 | block_size(raster, chunksize)
8 | }
9 | \arguments{
10 | \item{raster}{(\link[terra:SpatRaster-class]{terra::SpatRaster})\cr
11 | Raster to be split into chunks.}
12 |
13 | \item{chunksize}{(\code{integer(1)})\cr
14 | The chunksize determines in how many subparts the prediction task will be
15 | split into. The value can be roughly thought of as megabyte of a raster file
16 | on disk. For example, if a prediction on a 1 GB file would be carried out
17 | with \code{chunksize = 100L}, the prediction would happen in 10 chunks.
18 |
19 | The default of \code{chunksize = 1000L} might be a good compromise between speed
20 | and memory usage. If you find yourself running out of memory, reduce this
21 | value.}
22 | }
23 | \description{
24 | Splits raster into chunks.
25 | }
26 | \keyword{internal}
27 |
--------------------------------------------------------------------------------
/man/factor_layer.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/data.R
3 | \name{factor_layer}
4 | \alias{factor_layer}
5 | \title{Factor Layer Generator}
6 | \usage{
7 | factor_layer(id, levels, in_memory = FALSE)
8 | }
9 | \arguments{
10 | \item{id}{(\code{character(1)})\cr
11 | Layer id.}
12 |
13 | \item{levels}{(\code{character()})\cr
14 | Factor levels.}
15 |
16 | \item{in_memory}{(\code{logical(1)})\cr
17 | If \code{FALSE} (default), layer is written to disk.}
18 | }
19 | \value{
20 | Named \code{list()}
21 | }
22 | \description{
23 | Generates a factor layer when passed to \code{\link[=generate_stack]{generate_stack()}}.
24 | }
25 | \keyword{internal}
26 |
--------------------------------------------------------------------------------
/man/figures/land_cover.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/man/figures/land_cover.png
--------------------------------------------------------------------------------
/man/figures/logo.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/man/figures/logo.png
--------------------------------------------------------------------------------
/man/figures/sentinel.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/man/figures/sentinel.png
--------------------------------------------------------------------------------
/man/generate_stack.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/data.R
3 | \name{generate_stack}
4 | \alias{generate_stack}
5 | \title{Generate Raster Stack}
6 | \usage{
7 | generate_stack(
8 | layers,
9 | layer_size = NULL,
10 | dimension = NULL,
11 | multi_layer_file = FALSE
12 | )
13 | }
14 | \arguments{
15 | \item{layers}{(List of \code{\link[=numeric_layer]{numeric_layer()}} and \code{\link[=factor_layer]{factor_layer()}})\cr
16 | List of layers.}
17 |
18 | \item{layer_size}{(\code{numeric(1)})\cr
19 | Size of a single layer in megabytes.}
20 |
21 | \item{dimension}{(\code{integer(1)})\cr
22 | Dimension of the squared layers.}
23 |
24 | \item{multi_layer_file}{(\code{logical(1)})\cr
25 | If \code{TRUE}, raster is written to disk as a single multi-layer file.
26 | Overwrites \code{ìn_memory} argument of \code{numeric_layer()} and \code{factor_layer()}.
27 |
28 | \code{layer_size} and \code{dimension} are mutually exclusive.}
29 | }
30 | \value{
31 | \link[terra:SpatRaster-class]{terra::SpatRaster}
32 | }
33 | \description{
34 | Generates a raster stack.
35 | }
36 | \keyword{internal}
37 |
--------------------------------------------------------------------------------
/man/leipzig.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/TaskClassif_leipzig.R
3 | \docType{data}
4 | \name{leipzig}
5 | \alias{leipzig}
6 | \alias{mlr_tasks_leipzig}
7 | \title{Leipzig Land Cover Task}
8 | \source{
9 | Copernicus Sentinel Data (2021). Retrieved from Copernicus Open Access Hub and processed by European Space Agency.
10 | }
11 | \description{
12 | Point survey of land cover in Leipzig.
13 | Includes Sentinel-2 spectral bands and NDVI.
14 | }
15 | \examples{
16 | if (requireNamespace("sf")) {
17 | library(sf)
18 | data("leipzig", package = "mlr3spatial")
19 | print(leipzig)
20 | }
21 | }
22 | \keyword{data}
23 |
--------------------------------------------------------------------------------
/man/mask_stack.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/data.R
3 | \name{mask_stack}
4 | \alias{mask_stack}
5 | \title{Sample Points in Raster Stack}
6 | \usage{
7 | mask_stack(stack)
8 | }
9 | \arguments{
10 | \item{stack}{(\link[terra:SpatRaster-class]{terra::SpatRaster})\cr
11 | Raster stack.}
12 | }
13 | \value{
14 | \link[terra:SpatRaster-class]{terra::SpatRaster}
15 | }
16 | \description{
17 | Masks stack to a circular area of interest.
18 | }
19 | \keyword{internal}
20 |
--------------------------------------------------------------------------------
/man/mlr3spatial-package.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/zzz.R
3 | \docType{package}
4 | \name{mlr3spatial-package}
5 | \alias{mlr3spatial}
6 | \alias{mlr3spatial-package}
7 | \title{mlr3spatial: Support for Spatial Objects Within the 'mlr3' Ecosystem}
8 | \description{
9 | \if{html}{\figure{logo.png}{options: style='float: right' alt='logo' width='120'}}
10 |
11 | Extends the 'mlr3' ML framework with methods for spatial objects. Data storage and prediction are supported for packages 'terra', 'raster' and 'stars'.
12 | }
13 | \section{Learn mlr3}{
14 |
15 | \itemize{
16 | \item Book on mlr3: \url{https://mlr3book.mlr-org.com}
17 | \item Use cases and examples gallery: \url{https://mlr3gallery.mlr-org.com}
18 | \item Cheat Sheets: \url{https://github.com/mlr-org/mlr3cheatsheets}
19 | }
20 | }
21 |
22 | \section{mlr3 extensions}{
23 |
24 | \itemize{
25 | \item Preprocessing and machine learning pipelines: \CRANpkg{mlr3pipelines}
26 | \item Analysis of benchmark experiments: \CRANpkg{mlr3benchmark}
27 | \item More classification and regression tasks: \CRANpkg{mlr3data}
28 | \item Connector to \href{https://www.openml.org}{OpenML}: \CRANpkg{mlr3oml}
29 | \item Solid selection of good classification and regression learners: \CRANpkg{mlr3learners}
30 | \item Even more learners: \url{https://github.com/mlr-org/mlr3extralearners}
31 | \item Tuning of hyperparameters: \CRANpkg{mlr3tuning}
32 | \item Hyperband tuner: \CRANpkg{mlr3hyperband}
33 | \item Visualizations for many \pkg{mlr3} objects: \CRANpkg{mlr3viz}
34 | \item Survival analysis and probabilistic regression: \CRANpkg{mlr3proba}
35 | \item Cluster analysis: \CRANpkg{mlr3cluster}
36 | \item Feature selection filters: \CRANpkg{mlr3filters}
37 | \item Feature selection wrappers: \CRANpkg{mlr3fselect}
38 | \item Interface to real (out-of-memory) data bases: \CRANpkg{mlr3db}
39 | \item Performance measures as plain functions: \CRANpkg{mlr3measures}
40 | }
41 | }
42 |
43 | \section{Suggested packages}{
44 |
45 | \itemize{
46 | \item Parallelization framework: \CRANpkg{future}
47 | \item Progress bars: \CRANpkg{progressr}
48 | \item Encapsulated evaluation: \CRANpkg{evaluate}, \CRANpkg{callr} (external process)
49 | }
50 | }
51 |
52 | \section{Package Options}{
53 |
54 | \itemize{
55 | \item \code{"mlr3.debug"}: If set to \code{TRUE}, parallelization via \CRANpkg{future} is
56 | disabled to simplify debugging and provide more concise tracebacks. Note that
57 | results computed with debug mode enabled use a different seeding mechanism
58 | and are not reproducible.
59 | \item \code{"mlr3.allow_utf8_names"}: If set to \code{TRUE}, checks on the feature names
60 | are relaxed, allowing non-ascii characters in column names. This is an
61 | experimental and temporal option to pave the way for text analysis, and will
62 | likely be removed in a future version of the package. analysis.
63 | }
64 | }
65 |
66 | \references{
67 | Becker M, Schratz P (2024).
68 | \emph{mlr3spatial: Support for Spatial Objects Within the 'mlr3' Ecosystem}.
69 | https://mlr3spatial.mlr-org.com,
70 | https://github.com/mlr-org/mlr3spatial.
71 | }
72 | \seealso{
73 | Useful links:
74 | \itemize{
75 | \item \url{https://mlr3spatial.mlr-org.com}
76 | \item \url{https://github.com/mlr-org/mlr3spatial}
77 | \item Report bugs at \url{https://github.com/mlr-org/mlr3spatial/issues}
78 | }
79 |
80 | }
81 | \author{
82 | \strong{Maintainer}: Marc Becker \email{marcbecker@posteo.de} (\href{https://orcid.org/0000-0002-8115-0400}{ORCID})
83 |
84 | Authors:
85 | \itemize{
86 | \item Patrick Schratz \email{patrick.schratz@gmail.com} (\href{https://orcid.org/0000-0003-0748-6624}{ORCID})
87 | }
88 |
89 | }
90 |
--------------------------------------------------------------------------------
/man/numeric_layer.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/data.R
3 | \name{numeric_layer}
4 | \alias{numeric_layer}
5 | \title{Numeric Layer Generator}
6 | \usage{
7 | numeric_layer(id, in_memory = FALSE)
8 | }
9 | \arguments{
10 | \item{id}{(\code{character(1)})\cr
11 | Layer id.}
12 |
13 | \item{in_memory}{(\code{logical(1)})\cr
14 | If \code{FALSE} (default), layer is written to disk.}
15 | }
16 | \value{
17 | Named \code{list()}
18 | }
19 | \description{
20 | Generates a numeric layer when passed to \code{\link[=generate_stack]{generate_stack()}}.
21 | }
22 | \keyword{internal}
23 |
--------------------------------------------------------------------------------
/man/predict_spatial.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/predict_spatial.R
3 | \name{predict_spatial}
4 | \alias{predict_spatial}
5 | \title{Predict on Spatial Objects with mlr3 Learners}
6 | \usage{
7 | predict_spatial(
8 | newdata,
9 | learner,
10 | chunksize = 200L,
11 | format = "terra",
12 | filename = NULL
13 | )
14 | }
15 | \arguments{
16 | \item{newdata}{(\link[terra:SpatRaster-class]{terra::SpatRaster} | \code{stars::stars} | \link[sf:sf]{sf::sf} | \code{raster::RasterStack} | \code{raster::RasterBrick}).
17 | New data to predict on. All spatial data formats convertible by \code{as_data_backend()} are supported e.g. \link[terra:SpatRaster-class]{terra::SpatRaster} or \link[sf:sf]{sf::sf}.}
18 |
19 | \item{learner}{(\link{Learner}).
20 | Learner with trained model.}
21 |
22 | \item{chunksize}{(\code{integer(1)})\cr
23 | The chunksize determines in how many subparts the prediction task will be
24 | split into. The value can be roughly thought of as megabyte of a raster file
25 | on disk. For example, if a prediction on a 1 GB file would be carried out
26 | with \code{chunksize = 100L}, the prediction would happen in 10 chunks.
27 |
28 | The default of \code{chunksize = 1000L} might be a good compromise between speed
29 | and memory usage. If you find yourself running out of memory, reduce this
30 | value.}
31 |
32 | \item{format}{(\code{character(1)})\cr
33 | Output class of the resulting object.
34 | Accepted values are \code{"raster"}, \code{"stars"} and \code{"terra"} if the input is a raster.
35 | Note that when choosing something else than \code{"terra"}, the spatial object is converted into the respective format which might cause overhead both in runtime and memory allocation.
36 | For vector data only \code{"sf"} is supported.}
37 |
38 | \item{filename}{(\code{character(1)})\cr
39 | Path where the spatial object should be written to.}
40 | }
41 | \value{
42 | Spatial object of class given in argument \code{format}.
43 | }
44 | \description{
45 | This function allows to directly predict mlr3 learners on various spatial objects.
46 | }
47 | \examples{
48 | library(terra, exclude = "resample")
49 |
50 | # fit rpart on training points
51 | task_train = tsk("leipzig")
52 | learner = lrn("classif.rpart")
53 | learner$train(task_train)
54 |
55 | # load raster
56 | stack = rast(system.file("extdata", "leipzig_raster.tif", package = "mlr3spatial"))
57 |
58 | # predict land cover classes
59 | pred = predict_spatial(stack, learner, chunksize = 1L)
60 | }
61 |
--------------------------------------------------------------------------------
/man/sample_stack.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/data.R
3 | \name{sample_stack}
4 | \alias{sample_stack}
5 | \title{Sample Points in Raster Stack}
6 | \usage{
7 | sample_stack(stack, n = 100)
8 | }
9 | \arguments{
10 | \item{stack}{(\link[terra:SpatRaster-class]{terra::SpatRaster})\cr
11 | Raster stack.}
12 |
13 | \item{n}{(\code{integer(1)})\cr
14 | Number of points.}
15 | }
16 | \value{
17 | \link[sf:sf]{sf::sf}
18 | }
19 | \description{
20 | Samples \code{n} points of a raster stack.
21 | }
22 | \keyword{internal}
23 |
--------------------------------------------------------------------------------
/man/write_raster.Rd:
--------------------------------------------------------------------------------
1 | % Generated by roxygen2: do not edit by hand
2 | % Please edit documentation in R/helper.R
3 | \name{write_raster}
4 | \alias{write_raster}
5 | \title{Write a Raster in Chunks}
6 | \usage{
7 | write_raster(data)
8 | }
9 | \arguments{
10 | \item{data}{\verb{[SpatRaster]}\cr
11 | \code{SpatRaster} object.}
12 | }
13 | \description{
14 | Writes square raster to disk in chunks.
15 | Internal helper function.
16 | }
17 | \keyword{internal}
18 |
--------------------------------------------------------------------------------
/mlr3spatial.Rproj:
--------------------------------------------------------------------------------
1 | Version: 1.0
2 |
3 | RestoreWorkspace: Default
4 | SaveWorkspace: Default
5 | AlwaysSaveHistory: Default
6 |
7 | EnableCodeIndexing: Yes
8 | UseSpacesForTab: Yes
9 | NumSpacesForTab: 2
10 | Encoding: UTF-8
11 |
12 | RnwWeave: Sweave
13 | LaTeX: pdfLaTeX
14 |
15 | AutoAppendNewline: Yes
16 | StripTrailingWhitespace: Yes
17 |
18 | BuildType: Package
19 | PackageUseDevtools: Yes
20 | PackageCleanBeforeInstall: Yes
21 | PackageInstallArgs: --no-multiarch --with-keep.source
22 | PackageRoxygenize: rd,collate,namespace
23 |
--------------------------------------------------------------------------------
/pkgdown/_pkgdown.yml:
--------------------------------------------------------------------------------
1 | url: https://mlr3spatial.mlr-org.com
2 |
3 | template:
4 | bootstrap: 5
5 | package: mlr3pkgdowntemplate
6 |
7 | development:
8 | mode: auto
9 | version_label: default
10 | version_tooltip: "Version"
11 |
12 | authors:
13 | Patrick Schratz:
14 | href: https://pat-s.me
15 |
16 | navbar:
17 | structure:
18 | left: [reference, news, book, articles]
19 | right: [github, mattermost, stackoverflow, rss]
20 | components:
21 | home: ~
22 | reference:
23 | icon: fa fa-file-alt
24 | text: Reference
25 | href: reference/index.html
26 | mattermost:
27 | icon: fa fa-comments
28 | href: https://lmmisld-lmu-stats-slds.srv.mwn.de/mlr_invite/
29 | book:
30 | text: mlr3book
31 | icon: fa fa-link
32 | href: https://mlr3book.mlr-org.com
33 | stackoverflow:
34 | icon: fab fa-stack-overflow
35 | href: https://stackoverflow.com/questions/tagged/mlr3
36 | rss:
37 | icon: fa-rss
38 | href: https://mlr-org.com/
39 |
40 | reference:
41 | - title: mlr3spatial
42 | contents:
43 | - mlr3spatial-package
44 | - title: Spatiotemporal Tasks
45 | contents:
46 | - contains("Task")
47 | - contains("as_task_")
48 | - leipzig
49 | - title: Data Backends
50 | contents:
51 | - contains("Backend")
52 | - contains("_backend")
53 | - title: Prediction
54 | contents:
55 | - predict_spatial
56 |
--------------------------------------------------------------------------------
/pkgdown/apple-touch-icon-120x120.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/apple-touch-icon-120x120.png
--------------------------------------------------------------------------------
/pkgdown/apple-touch-icon-152x152.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/apple-touch-icon-152x152.png
--------------------------------------------------------------------------------
/pkgdown/apple-touch-icon-180x180.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/apple-touch-icon-180x180.png
--------------------------------------------------------------------------------
/pkgdown/apple-touch-icon-60x60.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/apple-touch-icon-60x60.png
--------------------------------------------------------------------------------
/pkgdown/apple-touch-icon-76x76.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/apple-touch-icon-76x76.png
--------------------------------------------------------------------------------
/pkgdown/apple-touch-icon.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/apple-touch-icon.png
--------------------------------------------------------------------------------
/pkgdown/favicon-16x16.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon-16x16.png
--------------------------------------------------------------------------------
/pkgdown/favicon-32x32.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon-32x32.png
--------------------------------------------------------------------------------
/pkgdown/favicon.ico:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon.ico
--------------------------------------------------------------------------------
/pkgdown/favicon/apple-touch-icon-120x120.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/apple-touch-icon-120x120.png
--------------------------------------------------------------------------------
/pkgdown/favicon/apple-touch-icon-152x152.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/apple-touch-icon-152x152.png
--------------------------------------------------------------------------------
/pkgdown/favicon/apple-touch-icon-180x180.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/apple-touch-icon-180x180.png
--------------------------------------------------------------------------------
/pkgdown/favicon/apple-touch-icon-60x60.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/apple-touch-icon-60x60.png
--------------------------------------------------------------------------------
/pkgdown/favicon/apple-touch-icon-76x76.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/apple-touch-icon-76x76.png
--------------------------------------------------------------------------------
/pkgdown/favicon/apple-touch-icon.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/apple-touch-icon.png
--------------------------------------------------------------------------------
/pkgdown/favicon/favicon-16x16.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/favicon-16x16.png
--------------------------------------------------------------------------------
/pkgdown/favicon/favicon-32x32.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/favicon-32x32.png
--------------------------------------------------------------------------------
/pkgdown/favicon/favicon.ico:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/mlr-org/mlr3spatial/f8e417455697159103f126ae89fc858a70861af8/pkgdown/favicon/favicon.ico
--------------------------------------------------------------------------------
/tests/testthat.R:
--------------------------------------------------------------------------------
1 | if (requireNamespace("testthat", quietly = TRUE)) {
2 | library("checkmate")
3 | library("testthat")
4 | library("mlr3spatial")
5 | test_check("mlr3spatial")
6 | }
7 |
--------------------------------------------------------------------------------
/tests/testthat/helper_expectations.R:
--------------------------------------------------------------------------------
1 | expect_backend = function(b) {
2 | checkmate::expect_r6(b, cloneable = FALSE,
3 | public = c("nrow", "ncol", "colnames", "rownames", "head", "data", "hash"),
4 | private = c(".data", ".hash", ".calculate_hash"))
5 | checkmate::expect_subset(b$data_formats, mlr3::mlr_reflections$data_formats, empty.ok = FALSE)
6 | testthat::expect_output(print(b), "^-
99 | #> 1 01-mlr3-4-cores 20.7s 21.4s 3 57 1.07m