diff --git a/src/raster/r.boxplot/r.boxplot.html b/src/raster/r.boxplot/r.boxplot.html index 20f6fd236c..e7fe2ee316 100644 --- a/src/raster/r.boxplot/r.boxplot.html +++ b/src/raster/r.boxplot/r.boxplot.html @@ -11,10 +11,10 @@DESCRIPTION
represent the median and interquartile range (IQR) of the input layer. Note that all values of the input raster (within the region's extent) are used to compute the median and IQR. If the zones of the zonal map -cover only part of the region, the user can mask out the non-covered +cover only part of the region, the user can mask out the non-covered parts of the input map first by means of r.mask. That will result in an IQR and median representing the values that fall within the zones of -the zonal map only. Otherwise, the computational region can be changed to +the zonal map only. Otherwise, the computational region can be changed to fit the extent of the zonal map with g.region@@ -25,7 +25,7 @@
DESCRIPTION
The whiskers extend to the most extreme data point, which is no -more than range ✕ the IQR from the box. +more than range ✕ the IQR from the box. By default, a range of 1.5 is used, but the user can change this. Note that range values need to be larger than 0. @@ -64,11 +64,11 @@
NOTE
with the error message, 'The zonal raster must be of type CELL (integer)'.-If the c flag is used, the bxp_color and median_color +If the c flag is used, the bxp_color and median_color are ignored, even if set by the user. The option to color boxploxs using the colors of the zonal raster categories (c flag) only works if the zonal map contains a color table. If it does not, the function exits with the -error message that 'The zonal map does not have a color table'. If the user +error message that 'The zonal map does not have a color table'. If the user thinks there is a color table, run r.colors.out and check if the categories are integers. If not, that is the problem. If they are all integers, you probably have caught a bug. @@ -83,7 +83,7 @@
EXAMPLE
Example 1
Draw a boxplot of the values of the elevation layer from the NC sample -dataset. Set the h flag to print the boxplot horizontally. +dataset. Set the h flag to print the boxplot horizontally. Set the plot dimensions to 7 inch wide, 1 inch high.diff --git a/src/raster/r.catchment/r.catchment.html b/src/raster/r.catchment/r.catchment.html index 18acd38da8..31403025d3 100644 --- a/src/raster/r.catchment/r.catchment.html +++ b/src/raster/r.catchment/r.catchment.html @@ -46,7 +46,7 @@Options and flags:
the catchment configuration. The optional value name_column is to be used in conjunction with the -i flag (see below). There are three native flags for r.catchment. -c -allows you to keep the interim cost surface maps made. -l allows +allows you to keep the interim cost surface maps made. -l allows you to show a list of the costv alues in that cost map, along with the size of the catchments they delineate. -i enable "iterative" mode. Here, the module will loop through all the points in the input @@ -72,7 +72,7 @@NOTES
determine the catchment. The input start_points map should be a vector points map. -If the file contains other types of features (areas, lines, centroids), +If the file contains other types of features (areas, lines, centroids), these will be ignored. If you desire, a start points map could be manually digitized (with v.digit) over topographic or cultural features, or could be created as a series of random points diff --git a/src/raster/r.edm.eval/r.edm.eval.html b/src/raster/r.edm.eval/r.edm.eval.html index 498b31f472..980f00ff22 100644 --- a/src/raster/r.edm.eval/r.edm.eval.html +++ b/src/raster/r.edm.eval/r.edm.eval.html @@ -1,29 +1,29 @@DESCRIPTION
-The r.edm.eval addon returns a few evaluation statistics, -including the area under the curve (AUC), the maximum true skill -statistic (TSS), and the maximum kappa. For all three statistics, the -corresponding probability threshold values are also given. Optionally, +The r.edm.eval addon returns a few evaluation statistics, +including the area under the curve (AUC), the maximum true skill +statistic (TSS), and the maximum kappa. For all three statistics, the +corresponding probability threshold values are also given. Optionally, it draws the receiver operating characteristic (ROC) curve.-The addon takes as input a raster layer with -observations(observations) and one or more -predictions layers. The observations layer -should be encoded with 1 (presence/cases) and 0 (absences/controls). -The (predictions layer gives the predicted values (e.g., the -outcome of a model). This typically are probability or suitability -scores between 0 and 1, but it will work on other scales too. With two -or more predictions layers, statistics are calculated for each -prediction layer separately. This allows one to compare predictions of +The addon takes as input a raster layer with +observations(observations) and one or more +predictions layers. The observations layer +should be encoded with 1 (presence/cases) and 0 (absences/controls). +The (predictions layer gives the predicted values (e.g., the +outcome of a model). This typically are probability or suitability +scores between 0 and 1, but it will work on other scales too. With two +or more predictions layers, statistics are calculated for each +prediction layer separately. This allows one to compare predictions of different models.
-By default, the prediction layer is divided in 200 bins (the user can -change this number). For each bin, the module computes the cumulative -true and false positives (TP, FP), true and false negatives (TN, FN), the -true and false positive rate (TPR, FPR), the true negative rate (TNR), -Cohen's kappa and the true skill statistic. +By default, the prediction layer is divided in 200 bins (the user can +change this number). For each bin, the module computes the cumulative +true and false positives (TP, FP), true and false negatives (TN, FN), the +true and false positive rate (TPR, FPR), the true negative rate (TNR), +Cohen's kappa and the true skill statistic.
TPR = TP / (TP + FN) @@ -42,49 +42,49 @@DESCRIPTION
-These values can be saved to a CSV file, together with the -corresponding upper and lower boundaries of the bins values. These can +These values can be saved to a CSV file, together with the +corresponding upper and lower boundaries of the bins values. These can be used to calculate additional statistics.
-With the -b flag, the module will use the fraction of -points that is predicted to be presence/true instead of the false -positive rate to create the ROC curve and calculate the AUC. Use this -when the predictions are created using a model that works with +With the -b flag, the module will use the fraction of +points that is predicted to be presence/true instead of the false +positive rate to create the ROC curve and calculate the AUC. Use this +when the predictions are created using a model that works with background points instead of (pseudo-)absence points (like Maxent).
-A problem with the AUC is that it varies with the spatial extent used -to select background points (Lobo et al. 2008, Jiménez-Valverde 2012). -Generally, the larger that extent, the higher the AUC value. Therefore, -AUC values are generally biased and cannot be directly compared. An -admittedly simple way to evaluate this issue of changing AUC values -with changing extents is to only use (pseudo-)absence or background -points within certain distances of the presence points (raster cells). -To do so, the user can use the buffer parameter to limit the -absence/background points that are used to calculate the various -statistics to those within a certain distance from the presence -locations/areas. +A problem with the AUC is that it varies with the spatial extent used +to select background points (Lobo et al. 2008, Jiménez-Valverde 2012). +Generally, the larger that extent, the higher the AUC value. Therefore, +AUC values are generally biased and cannot be directly compared. An +admittedly simple way to evaluate this issue of changing AUC values +with changing extents is to only use (pseudo-)absence or background +points within certain distances of the presence points (raster cells). +To do so, the user can use the buffer parameter to limit the +absence/background points that are used to calculate the various +statistics to those within a certain distance from the presence +locations/areas.
-The user can set the required ratio of presence and total points to be -used in the evaluation (preval; value between 0 an 1). This -allows the user to test how sensitive the evaluation statistics are for -the prevalence of presence points. +The user can set the required ratio of presence and total points to be +used in the evaluation (preval; value between 0 an 1). This +allows the user to test how sensitive the evaluation statistics are for +the prevalence of presence points.
NOTES
-The module expects an reference raster layer, encoded with 1 -(presence/cases) and 0 (absences/controls). Optionally, the user can -define other values using the absence and presence -parameters. If the layer contains more than 2 unique values, or if the -layer is not of the type CELL (integer), the module will terminate with +The module expects an reference raster layer, encoded with 1 +(presence/cases) and 0 (absences/controls). Optionally, the user can +define other values using the absence and presence +parameters. If the layer contains more than 2 unique values, or if the +layer is not of the type CELL (integer), the module will terminate with a warning.-The selection of the best evaluation statistic depends on your data and -what you want to test. This module provides a few only, but the user can -calculate his/her own using the table in the CSV file. See the +The selection of the best evaluation statistic depends on your data and +what you want to test. This module provides a few only, but the user can +calculate his/her own using the table in the CSV file. See the References list for a few useful references in that respect.
@@ -93,9 +93,9 @@
NOTES
EXAMPLES
-Run this example in the "landsat" mapset of the North Carolina sample -data set location. First, create a binary map with herbaceous land cover -(1) and other (0). Next, create a raster layer with training pixels +Run this example in the "landsat" mapset of the North Carolina sample +data set location. First, create a binary map with herbaceous land cover +(1) and other (0). Next, create a raster layer with training pixels using r.learn.train and r.learn.predict from the r.learn.ml2 addon. Note, r.learn.train can compute its own accuracy @@ -108,7 +108,7 @@EXAMPLES
-Then we can use these training pixels to perform a classification on +Then we can use these training pixels to perform a classification on the more recently obtained landsat 7 image using r.learn.train.
@@ -116,14 +116,14 @@EXAMPLES
r.learn.train group=lsat7_2000 training_map=training_pixels \ model_name=RandomForestRegressor n_estimators=500 \ save_model=rf_model.gz cv=2 - + # perform prediction using r.learn.predict r.learn.predict group=lsat7_2000 load_model=rf_model.gz \ output=rf_regressor-Now we can see how well the model performed using the +Now we can see how well the model performed using the r.edm.eval.
@@ -154,14 +154,14 @@EXAMPLES
-Let's try how well a logistic regression performs, and compare this +Let's try how well a logistic regression performs, and compare this with the previous results.
-And the accompanying ROC curve is shown below (the figure can optionally be saved to file). +And the accompanying ROC curve is shown below (the figure can optionally be saved to file).# train a logistic regression model using r.learn.train r.learn.train group=lsat7_2000 training_map=training_pixels \ model_name=LogisticRegression save_model=lr_model.gz cv=2 - + # perform prediction using r.learn.predict -p r.learn.predict group=lsat7_2000 load_model=lr_model.gz \ output=lr_regressor @@ -191,7 +191,7 @@EXAMPLES
treshold minimum distance to (0,1) = 0.0848
@@ -204,29 +204,29 @@EXAMPLES
REFERENCES
Jiménez-Valverde, A. 2012. Insights into the area under the receiver -operating characteristic curve (AUC) as a discrimination measure in -species distribution modelling. Global Ecology and Biogeography 21: +operating characteristic curve (AUC) as a discrimination measure in +species distribution modelling. Global Ecology and Biogeography 21: 498-507Lobo, J. M., Jiménez-Valverde, A., & Real, R. 2008. AUC: a -misleading measure of the performance of predictive distribution +misleading measure of the performance of predictive distribution models. Global Ecology and Biogeography 17: 145-151.
Hijmans, R. J. 2012. Cross-validation of species distribution models: -removing spatial sorting bias and calibration with a null model. +removing spatial sorting bias and calibration with a null model. Ecology 93: 679-688.
Allouche, O., Tsoar, A., & Kadmon, R. 2006. Assessing the accuracy of -species distribution models: prevalence, kappa and the true skill +species distribution models: prevalence, kappa and the true skill statistic (TSS). Journal of Applied Ecology 43: 1223-1232.
McPherson, J. M., Jetz, W., & Rogers, D. J. 2004. The effects of -species' range sizes on the accuracy of distribution models: ecological -phenomenon or statistical artefact? Journal of Applied Ecology 41: +species' range sizes on the accuracy of distribution models: ecological +phenomenon or statistical artefact? Journal of Applied Ecology 41: 811-823.
SEE ALSO
diff --git a/src/raster/r.fusion/r.fusion.html b/src/raster/r.fusion/r.fusion.html index 59c1f5ea14..a7d2834292 100644 --- a/src/raster/r.fusion/r.fusion.html +++ b/src/raster/r.fusion/r.fusion.html @@ -1,15 +1,15 @@DESCRIPTION
-r.fusion enhances the resolution of a raster map by using spatial -detail of a high-resolution map. The actual values in the resultant output map -correspond to the input map while the spatial detail corresponds to the -high-resolution map. The effect is similar to pan-sharpening, but the method -can be applied more generally, not only to imagery but also to climatological +r.fusion enhances the resolution of a raster map by using spatial +detail of a high-resolution map. The actual values in the resultant output map +correspond to the input map while the spatial detail corresponds to the +high-resolution map. The effect is similar to pan-sharpening, but the method +can be applied more generally, not only to imagery but also to climatological data such as temperature or precipitation.NOTES
-Two different methods are available with the method option: +Two different methods are available with the method option: difference and proportion.@@ -21,11 +21,11 @@
NOTES
highres(Alowres - Blowres) + Bhighres = Ahighres
-where highres() is a function to interpolate the differences. Here, +where highres() is a function to interpolate the differences. Here, r.resamp.filter is used for interpolation.
-The proportion method is suitable for e.g. precipitation where zero +The proportion method is suitable for e.g. precipitation where zero precipition must stay zero precipition, and uses the formula
A / B * B = A @@ -35,8 +35,8 @@
NOTES
highres(Alowres / Blowres) * Bhighres = Ahighres-Again, highres() is a function to interpolate the proportions, and -r.resamp.filter is used for interpolation. For the proportion +Again, highres() is a function to interpolate the proportions, and +r.resamp.filter is used for interpolation. For the proportion method, all values in the high-resolution B map must be > 0. diff --git a/src/raster/r.fuzzy.set/r.fuzzy.set.html b/src/raster/r.fuzzy.set/r.fuzzy.set.html index f2e554c533..eefd41d2b3 100644 --- a/src/raster/r.fuzzy.set/r.fuzzy.set.html +++ b/src/raster/r.fuzzy.set/r.fuzzy.set.html @@ -93,7 +93,7 @@
Calculation of boundary shape
1-cos(x * Pi/2)^m (for negative shape parameter) where x: membership, and -m = 2^exp(2 * |shape|) +m = 2^exp(2 * |shape|) For default shape=0, m = 2 (most common parameter for that equation). diff --git a/src/raster/r.in.ahn/r.in.ahn.html b/src/raster/r.in.ahn/r.in.ahn.html index 6fb82f5aa2..5cd4b94d42 100644 --- a/src/raster/r.in.ahn/r.in.ahn.html +++ b/src/raster/r.in.ahn/r.in.ahn.html @@ -1,48 +1,48 @@DESCRIPTION
-r.in.ahn imports the Actueel Hoogtebestand Nederland (AHN, version 4) for the current region. -The AHN is a digital elevation model (DEM) of the Netherlands with a -resolution of of 0.5 meter. +r.in.ahn imports the Actueel Hoogtebestand Nederland (AHN, version 4) for the current region. +The AHN is a digital elevation model (DEM) of the Netherlands with a +resolution of of 0.5 meter.-There are two different layers available: the digital terrain model -(DTM) and a digital surface model (DSM). The user needs to select which -to download. The selected product will be downloaded for the -computation region. However, note that the region will adjusted to -ensure that the imported raster layer neatly aligns with and has the -same resolution (0.5 meter) as the original AHN data. The resulting -will always have the same or a larger extent than the original -computation region. If you want to store the current computational -region, make sure to first save it using g.region. +There are two different layers available: the digital terrain model +(DTM) and a digital surface model (DSM). The user needs to select which +to download. The selected product will be downloaded for the +computation region. However, note that the region will adjusted to +ensure that the imported raster layer neatly aligns with and has the +same resolution (0.5 meter) as the original AHN data. The resulting +will always have the same or a larger extent than the original +computation region. If you want to store the current computational +region, make sure to first save it using g.region.
-The AHN can also be downloaded in map sheets (tiles) of 6.5 by 5 -kilometer. To download the area covered by one or more of these tiles, -the user can set the -t flag. This wil to download the area for +The AHN can also be downloaded in map sheets (tiles) of 6.5 by 5 +kilometer. To download the area covered by one or more of these tiles, +the user can set the -t flag. This wil to download the area for the tiles that overlap with the current computational region.
NOTE
- -This location only works in a location with the project 'RD New' -(EPSG:28992). Attempts to run it in a location with another CRS will + +This location only works in a location with the project 'RD New' +(EPSG:28992). Attempts to run it in a location with another CRS will result in an error message.-The region will be adjusted to ensure that the imported raster layer -neatly aligns with and has the same resolution (0.5 meter) as the -original AHN data. The user can set the -g flag to return the +The region will be adjusted to ensure that the imported raster layer +neatly aligns with and has the same resolution (0.5 meter) as the +original AHN data. The user can set the -g flag to return the region to the original computation region after the data is imported.
-The addon uses the r.in.wcs addon in the background, so the +The addon uses the r.in.wcs addon in the background, so the user will first need to install this addon.
EXAMPLE
Example 1
-Import the DTM for Fort Crèvecoeur, an fortress where the river Old +Import the DTM for Fort Crèvecoeur, an fortress where the river Old Dieze flows into the Maas river.@@ -64,7 +64,7 @@
Example 1
Example 2
-Import the DTM for the tile that overlaps with the current region. Do +Import the DTM for the tile that overlaps with the current region. Do this by setting the -t flag.@@ -77,7 +77,7 @@
Example 2
-The result will be a raster layer dsm_crevecoeur2 and a vector +The result will be a raster layer dsm_crevecoeur2 and a vector layer dsm_crevecoeur2_tiles
@@ -85,20 +85,20 @@
Example 2
-Figure: DSM map of Fort Crèvecoeur. Left shows the extent (red -outline) after running example 2. The extent equals the extent of the -area (tile) for which the data was downloaded. Right shows the extent -(red outline) after running example 3. In this case, the extent is the -same as before running the example because the -g flag was +Figure: DSM map of Fort Crèvecoeur. Left shows the extent (red +outline) after running example 2. The extent equals the extent of the +area (tile) for which the data was downloaded. Right shows the extent +(red outline) after running example 3. In this case, the extent is the +same as before running the example because the -g flag was set.
-
-