Merge pull request #477 from e-sensing/dev

Pre-release 0.15.1
e-sensing · Nov 17, 2021 · edb09d3 · edb09d3
2 parents c92a3b7 + b68ca8c
commit edb09d3
Show file tree

Hide file tree

Showing 215 changed files with 2,926 additions and 9,059 deletions.
diff --git a/.gitignore b/.gitignore
@@ -26,3 +26,5 @@ Makevars
 Makevars.*
 *.gcda
 *.gcno
+^_snaps
+
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,6 +1,6 @@
 Package: sits
 Type: Package
-Version: 0.15.0-4
+Version: 0.15.1
 Title: Satellite Image Time Series Analysis for Remote Sensing Data Cubes
 Authors@R: c(person('Rolf', 'Simoes', role = c('aut'), email = 'rolf.simoes@inpe.br'),
              person('Gilberto', 'Camara', role = c('aut', 'cre'), email = 'gilberto.camara@inpe.br'),
@@ -43,7 +43,8 @@ Imports:
     parallel,
     purrr (>= 0.3.0),
     raster (>= 3.4),
-    rstac (>= 0.9.1-4),
+    Rcpp (>= 1.0.7),
+    rstac (>= 0.9.1-5),
     sf (>= 0.9),
     slider (>= 0.1.5),
     stats,
@@ -67,7 +68,6 @@ Suggests:
     nnet,
     openxlsx,
     randomForest,
-    Rcpp,
     RcppArmadillo,
     Rwtss (>= 0.9.1),
     scales,
@@ -90,7 +90,6 @@ Collate:
     'sits_accuracy.R'
     'sits_bands.R'
     'sits_bbox.R'
-    'sits_brewer.R'
     'sits_classification.R'
     'sits_classify_ts.R'
     'sits_classify_cube.R'
@@ -107,13 +106,10 @@ Collate:
     'sits_filters.R'
     'sits_gdalcubes.R'
     'sits_get_data.R'
-    'sits_indexes.R'
     'sits_imputation.R'
     'sits_keras.R'
     'sits_labels.R'
-    'sits_labels_set.R'
     'sits_label_classification.R'
-    'sits_labels_summary.R'
     'sits_machine_learning.R'
     'sits_merge.R'
     'sits_mlp.R'
@@ -144,6 +140,7 @@ Collate:
     'sits_source_api_bdc.R'
     'sits_source_api_deafrica.R'
     'sits_source_api_local.R'
+    'sits_source_api_mspc.R'
     'sits_source_api_satveg.R'
     'sits_source_api_stac.R'
     'sits_source_api_usgs.R'

diff --git a/NAMESPACE b/NAMESPACE
@@ -1,6 +1,6 @@
 # Generated by roxygen2: do not edit by hand
 
-S3method("sits_labels<-",pattern)
+S3method("sits_labels<-",probs_cube)
 S3method("sits_labels<-",sits)
 S3method(.raster_check_package,default)
 S3method(.raster_check_package,raster)
@@ -58,6 +58,7 @@ S3method(.raster_yres,terra)
 S3method(.sits_roi_bbox,ll)
 S3method(.sits_roi_bbox,sf)
 S3method(.sits_roi_bbox,xy)
+S3method(.source_collection_access_test,mspc_cube)
 S3method(.source_collection_access_test,satveg_cube)
 S3method(.source_collection_access_test,stac_cube)
 S3method(.source_collection_access_test,wtss_cube)
@@ -75,19 +76,26 @@ S3method(.source_items_cube,stac_cube)
 S3method(.source_items_cube,wtss_cube)
 S3method(.source_items_fileinfo,stac_cube)
 S3method(.source_items_fileinfo,wtss_cube)
+S3method(.source_items_new,"mspc_cube_landsat-8-c2-l2")
+S3method(.source_items_new,"mspc_cube_sentinel-2-l2a")
 S3method(.source_items_new,aws_cube)
 S3method(.source_items_new,bdc_cube)
 S3method(.source_items_new,deafrica_cube)
 S3method(.source_items_new,usgs_cube)
 S3method(.source_items_new,wtss_cube)
 S3method(.source_items_tile_get_bbox,stac_cube)
+S3method(.source_items_tile_get_crs,"mspc_cube_landsat-8-c2-l2")
+S3method(.source_items_tile_get_crs,"mspc_cube_sentinel-2-l2a")
 S3method(.source_items_tile_get_crs,aws_cube)
 S3method(.source_items_tile_get_crs,bdc_cube)
 S3method(.source_items_tile_get_crs,deafrica_cube)
 S3method(.source_items_tile_get_crs,usgs_cube)
+S3method(.source_items_tile_get_name,"mspc_cube_sentinel-2-l2a")
 S3method(.source_items_tile_get_name,bdc_cube)
 S3method(.source_items_tile_get_name,deafrica_cube)
 S3method(.source_items_tile_get_name,stac_cube)
+S3method(.source_items_tiles_group,"mspc_cube_landsat-8-c2-l2")
+S3method(.source_items_tiles_group,"mspc_cube_sentinel-2-l2a")
 S3method(.source_items_tiles_group,aws_cube)
 S3method(.source_items_tiles_group,bdc_cube)
 S3method(.source_items_tiles_group,deafrica_cube)
@@ -145,6 +153,7 @@ S3method(sits_timeline,probs_cube)
 S3method(sits_timeline,raster_cube)
 S3method(sits_timeline,satveg_cube)
 S3method(sits_timeline,sits)
+S3method(sits_timeline,sits_model)
 S3method(sits_timeline,wtss_cube)
 S3method(sits_values,bands_cases_dates)
 S3method(sits_values,bands_dates_cases)
@@ -168,42 +177,32 @@ export(sits_cluster_dendro)
 export(sits_cluster_frequency)
 export(sits_config)
 export(sits_config_show)
-export(sits_create_folds)
 export(sits_cube)
-export(sits_data_to_csv)
-export(sits_envelope)
 export(sits_filter)
 export(sits_formula_linear)
 export(sits_formula_logref)
-export(sits_from_zoo)
 export(sits_get_data)
 export(sits_impute_linear)
 export(sits_interp)
-export(sits_keras_diagnostics)
 export(sits_kfold_validate)
 export(sits_label_classification)
 export(sits_labels)
 export(sits_labels_summary)
-export(sits_linear_interp)
 export(sits_list_collections)
 export(sits_merge)
 export(sits_metadata_to_csv)
-export(sits_missing_values)
 export(sits_mlp)
 export(sits_mlr)
 export(sits_mutate_bands)
-export(sits_ndwi)
 export(sits_patterns)
 export(sits_regularize)
 export(sits_rfor)
 export(sits_sample)
-export(sits_savi)
 export(sits_select)
 export(sits_sgolay)
 export(sits_show_prediction)
 export(sits_smooth)
 export(sits_som_clean_samples)
-export(sits_som_cluster)
 export(sits_som_evaluate_cluster)
 export(sits_som_map)
 export(sits_svm)
@@ -217,6 +216,7 @@ export(sits_values)
 export(sits_view)
 export(sits_whittaker)
 export(sits_xgboost)
+importFrom(Rcpp,sourceCpp)
 importFrom(data.table,":=")
 importFrom(lubridate,"%m+%")
 importFrom(lubridate,"%within%")

diff --git a/NEWS.md b/NEWS.md
@@ -2,6 +2,30 @@
 
 We are preparing to release the package on CRAN and are making relevant changes in the development version.
 
+# What's new in SITS version 0.15.1
+
+### New features in SITS version 0.15.1
+* Update README.Rmd
+* Support `sits_timeline()` to sits model objects
+* Update drone image
+* Simplify `config_colors.yml` by removing palette names
+* Temporary python files are being generated in the check
+* Organise color handling in SITS
+* Organise configuration files
+* Improve preconditions in `sits_regularize()`
+* Compress external data with bzip2
+* Update gdalcubes format files
+* Update rstac version
+* Check provided parameters in sits_regularize function
+* Use default palette for SOM colors
+* Remove `start_date` and `end_date` from validation csv file
+* Use a default brewer palette to plot classified cube
+* Improve package help pages
+* Remove unused data sets
+* Remove rarely used functions
+* `sits_regularize()` is producing *Float64* images as output
+* Full support for Microsoft Planetary Computing
+
 # What's new in SITS version 0.15.0 
 
 ### New features in SITS version 0.15.0-4

diff --git a/R/data.R b/R/data.R
@@ -57,64 +57,3 @@ NULL
 #'   cube (the name of the cube associated with the data),
 #'   time_series (list containing a tibble with the values of the time series).
 NULL
-
-#' @title Samples of nine classes for the state of Mato Grosso
-#'
-#' @description A dataset containing a tibble with time series samples
-#'   for the Mato Grosso state in Brasil.
-#'   The time series come from MOD13Q1 collection 6 images.
-#'   The data set has the following classes:
-#'   Cerrado, Fallow_Cotton, Forest, Pasture,
-#'   Soy_Corn, Soy_Cotton, Soy_Fallow,
-#'   Soy_Millet, and Soy_Sunflower.
-#'
-#' @docType data
-#' @keywords datasets
-#' @name samples_mt_6bands
-#' @usage data(samples_mt_6bands)
-#' @format A tibble with 425 rows and 7 variables:
-#'   longitude: East-west coordinate of the time series sample (WGS 84),
-#'   latitude (North-south coordinate of the time series sample in WGS 84),
-#'   start_date (initial date of the time series),
-#'   end_date (final date of the time series),
-#'   label (the class label associated to the sample),
-#'   cube (the name of the cube associated with the data),
-#'   time_series (list containing a tibble with the values of the time series).
-NULL
-
-#' @title A time series in the ZOO format
-#'
-#' @description A dataset containing a one ZOO time series samples
-#'    in the Mato Grosso state of Brazil.
-#'    The time series comes from MOD13Q1 collection 5 images.
-#'
-#' @docType data
-#' @keywords datasets
-#' @name ts_zoo
-#' @usage data(ts_zoo)
-#' @format A ZOO time series with bands NDVI and EVI for a one year period
-NULL
-
-#' @title The timeline for the sequence of images for MOD13Q1 collection 6
-#'
-#' @description The timeline for the time series used in the examples
-#'    that use data from MODIS collection 6.
-#'    There are 407 instances from 2000-02-18 until 2017-10-16.
-#' @docType data
-#' @keywords datasets
-#' @name timeline_2000_2017
-#' @usage data(timeline_2000_2017)
-#' @format A vector with 407 dates in YYYY-MM-DD format.
-NULL
-
-#' @title The timeline for the sequence of images one year (2013 to 2014)
-#'
-#' @description The timeline for the time series used in the examples
-#'    that use data for one year classification.
-#'    There are 23 instances from 2013-09-14 until 2014-08-29.
-#' @docType data
-#' @keywords datasets
-#' @name timeline_2013_2014
-#' @usage data(timeline_2013_2014)
-#' @format A vector with 23 dates in YYYY-MM-DD format.
-NULL
diff --git a/R/sits-package.R b/R/sits-package.R
@@ -2,7 +2,7 @@
 #' @description Satellite Image Time Series Analysis
 #'              for Earth Observation Data Cubes
 #'
-#' @section SITS API:
+#' @section Purpose:
 #'
 #' The SITS package provides a set of tools for analysis,
 #' visualization and classification of satellite image time series.

diff --git a/R/sits_ResNet.R b/R/sits_ResNet.R
@@ -1,4 +1,4 @@
-#' @title Train a model using the ResNet model
+#' @title Train ResNet classification models
 #' @name sits_ResNet
 #'
 #' @author Gilberto Camara, \email{gilberto.camara@@inpe.br}
@@ -96,12 +96,10 @@ sits_ResNet <- function(samples = NULL,
             stop("Please install package keras", call. = FALSE)
         }
 
-        valid_activations <- c("relu", "elu", "selu", "sigmoid")
-        # pre-conditions
         .check_chr_within(
             x = activation,
-            within = valid_activations,
-            discriminator = "any_of",
+            within = .config_get("dl_activation_methods"),
+            discriminator = "one_of",
             msg = "invalid CNN activation method"
         )
 

diff --git a/R/sits_TempCNN.R b/R/sits_TempCNN.R
@@ -1,5 +1,5 @@
 
-#' @title Train a model using the Temporal Convolutional Neural Network
+#' @title Train temporal convolutional neural network models
 #' @name sits_TempCNN
 #'
 #' @author Gilberto Camara, \email{gilberto.camara@@inpe.br}
@@ -95,9 +95,6 @@ sits_TempCNN <- function(samples = NULL,
             stop("Please install package keras", call. = FALSE)
         }
 
-        # pre-conditions
-        valid_activations <- c("relu", "elu", "selu", "sigmoid")
-
         .check_length(
             x = cnn_layers,
             len_min = length(cnn_kernels),
@@ -124,14 +121,14 @@ sits_TempCNN <- function(samples = NULL,
 
         .check_chr_within(
             x = cnn_activation,
-            within = valid_activations,
+            within = .config_get("dl_activation_methods"),
             discriminator = "one_of",
             msg = "invalid CNN activation method"
         )
 
         .check_chr_within(
             x = dense_layer_activation,
-            within = valid_activations,
+            within = .config_get("dl_activation_methods"),
             discriminator = "one_of",
             msg = "invalid node activation method"
         )

diff --git a/R/sits_accuracy.R b/R/sits_accuracy.R
@@ -1,4 +1,4 @@
-#' @title Area-weighted classification accuracy assessment
+#' @title Assess classification accuracy (area-weighted method)
 #' @name sits_accuracy
 #' @author Rolf Simoes, \email{rolf.simoes@@inpe.br}
 #' @author Alber Sanchez, \email{alber.ipia@@inpe.br}
@@ -167,7 +167,10 @@ sits_accuracy.classified_image <- function(data, ..., validation_csv) {
                                                 stringsAsFactors = FALSE))
 
     # Precondition - check if CSV file is correct
-    .sits_csv_check(csv_tb)
+    .check_chr_contains(
+        x = colnames(csv_tb),
+        contains = c("longitude", "latitude", "label"),
+        msg = "invalid csv file")
 
     # find the labels of the cube
     labels_cube <- sits_labels(data)
@@ -204,8 +207,7 @@ sits_accuracy.classified_image <- function(data, ..., validation_csv) {
         points_row <- dplyr::filter(
             points,
             X >= row$xmin & X <= row$xmax &
-                Y >= row$ymin & Y <= row$ymax,
-            start_date == row$file_info[[1]]$start_date
+                Y >= row$ymin & Y <= row$ymax
         )
 
         # if there are no points in the cube, return an empty list

diff --git a/R/sits_bands.R b/R/sits_bands.R
@@ -1,11 +1,11 @@
-#' @title Informs the names of the bands
+#' @title Get the names of the bands
 #'
 #' @name sits_bands
 #'
 #' @author Gilberto Camara, \email{gilberto.camara@@inpe.br}
 #' @author Rolf Simoes, \email{rolf.simoes@@inpe.br}
 #'
-#' @description  Finds/replaces the names of the bands of
+#' @description  Finds the names of the bands of
 #'               a set of time series or of a data cube
 #'
 #' @param x         Valid sits tibble (time series or a cube)
@@ -15,7 +15,7 @@
 #' @examples {
 #' # Retrieve the set of samples for Mato Grosso (provided by EMBRAPA)
 #' # show the bands
-#' sits_bands(samples_mt_6bands)
+#' sits_bands(samples_modis_4bands)
 #' }
 #'
 #' @export