rasterly (version 0.2.0)

rasterly_points: rasterly_points

Description

Points layer for rasterly.

Usage

rasterly_points(
  rastObj,
  data = NULL,
  mapping = aes(),
  ...,
  xlim = NULL,
  ylim = NULL,
  max_size = NULL,
  reduction_func = NULL,
  layout = NULL,
  glyph = NULL,
  group_by_data_table = NULL,
  inherit.aes = TRUE
)

Arguments

rastObj

A rasterly object.

data

A data.frame or function with an argument x, specifying the dataset to use for plotting. If data is NULL, the data argument provided to rasterly may be passed through.

mapping

Default list of aesthetic mappings to use for plot. If provided and inherit.aes = TRUE, it will be stacked on top of the mappings passed to rasterly.

...

Pass-through arguments provided by rasterly.

xlim

Vector of type numeric. X limits in this layer.

ylim

Vector of type numeric. Y limits in this layer.

max_size

Numeric. When size changes, the upper bound of the number of pixels over which to spread a single observation.

reduction_func

Function. A reduction function is used to aggregate data points into their pixel representations. Currently supported reduction operators are sum, any, mean, m2, first, last, min and max. Default is sum. See details.

layout

Character. The method used to generate layouts for multiple images. The default is weighted. Useful for categorical data (i.e. "color" is provided via aes()). weighted specifies that the final raster should be a weighted combination of each (categorical) aggregation matrix. Conversely, cover indicates that the afterwards objects will be drawn on top of the previous ones.

glyph

Character. Currently, only "circle" and "square" are supported; as the size of the pixels increases, how should they spread out -- should the pattern be circular or square? Other glyphs may be added in the future.

group_by_data_table

Logical. Default is TRUE; when "color" is provided via aes(), the "group by" operation may be perfromed within data.table or natively within rasterly. Generally, group_by_data_table = TRUE is faster, but for very large datasets grouping within rasterly may offer better performance.

inherit.aes

If FALSE, overrides the default aesthetics, rather than combining with them.

Value

A list of environments.

Details

Reduction functions

• sum: If on is not provided within aes(), the default is to take the sum within each bin. When on is specified, the function reduces by taking the sum of all elements within the variable named in on.

• any: When on is provided within aes(), the any reduction function specifies whether any elements in on should be mapped to each bin.

• mean: If on is not provided in mapping aes(), on would be set as variable "y" by default. When on is given, the mean reduction function takes the mean of all elements within the variable specified by on.

• m2: Requires that on is specified within aes(). The m2 function computes the sum of square differences from the mean of all elements in the variable specified by on.

• var: Requires that on is specified within aes(). The var function computes the variance over all elements in the vector specified by on.

• sd: Requires that on is specified within aes(). The sd function computes the standard deviation over all elements in the vector specified by on.

• first: Requires that on is specified within aes(). The first function returns the first element in the vector specified by on.

• last: Requires that on is specified within aes(). The last function returns the last element in the vector specified by on.

• min: Requires that on is specified within aes(). The min function returns the minimum value in the vector specified by on.

• max: Requires that on is specified within aes(). The min function returns the maximum value in the vector specified by on.

See Also

rasterly, rasterly_build, [.rasterly, [<-.rasterly

Examples

# NOT RUN {
   library(rasterly)
   if(requireNamespace("grid") && requireNamespace("gridExtra")) {
     x <- rnorm(1e7)
     y <- rnorm(1e7)
     category <- sample(1:5, 1e7, replace = TRUE)
     data.frame(x = x, y = y, category = category) %>%
       rasterly(mapping = aes(x = x, y = y, color = category)) %>%
       rasterly_points(layout = "weighted") -> ds1
     ds1
     # layout with cover
     data.frame(x = x, y = y, category = category) %>%
       rasterly(mapping = aes(x = x, y = y, color = category)) %>%
       rasterly_points(layout = "cover") -> ds2
     ds2
     # display side by side
     grid::grid.newpage()
     gridExtra::grid.arrange(
        grobs = list(rasterlyGrob(ds1), rasterlyGrob(ds2)),
        ncol = 2,
        top = "'weighted' layout versus 'cover' layout"
     )
   }
# }