Skip to contents

Set and retrieve BirdFlowR options

Source: R/birdflow_options.R
birdflow_options.Rd

With no arguments all the BirdFlowR options will be returned as a list. Use a single character value to retrieve the value of a single option. Use one or more named arguments to set options.

Usage

birdflow_options(...)

Arguments

...

One of: (1) one or more named arguments where the name is a an option and the value its new setting e.g. verbose = FALSE ; (2) a single unnamed argument stating an option to retrieve e.g. "verbose" with an option to retrieve. (3) No arguments, indicating that all options and their current settings should be returned in a list; or (4) a single list argument with named items and their new values.

Value

If no arguments are used than all options will be returned as a list. If there is a single, unnamed argument with a character value indicating an option than the value of that option will be returned. Otherwise, the arguments should indicate new option settings and nothing will be returned.

Details

time_format

Indicates what time format to use to label dimensions of distribution tables and layers of raster objects returned by get_distr(), rast(), rasterize_distr(), and predict(). It does not affect internally stored distribution column labels (which are always t1, t2, etc.). Default is "month_day".

Valid values are: "timestep" which uses the timestep integer appended to "t" e.g. "t1"); "date" which uses a date in the format year-month-day (as numbers) e.g. "2022-11-23"; and "month_day" which uses the name of the month followed by the day of the month e.g. "November 23."

verbose

Defaults to TRUE for printing of progress and information about the process of functions. Set to FALSE to turn off printing.

max_param_per_gpu_gb

Controls how many parameters can be fit by BirdFlowPy per gigabyte of GPU Ram. This is a conservative estimate based on empirical testing. See preprocess_species()

cache

The local directory to store downloaded model files. Defaults to tools::R_user_dir("BirdFlowR", "data"). This is the base cache directory within which there will be one or more collection specific directories, which in turn will hold BirdFlow model files and an index.

collection_url

This is the base URL of a collection of model files and its associated index. The default is for the standard BirdFlowR model collection.

Examples

bf <- BirdFlowModels::amewoo
birdflow_options() # print current settings
#> $cache
#> [1] "/home/runner/.local/share/R/BirdFlowR"
#> 
#> $collection_url
#> [1] "https://birdflow-science.s3.amazonaws.com/collection/"
#> 
#> $max_param_per_gpu_gb
#> [1] 23224801
#> 
#> $time_format
#> [1] "month_day"
#> 
#> $verbose
#> [1] TRUE
#> 
original_format <- birdflow_options("time_format")
birdflow_options(time_format = "date")
head(get_distr(bf, 1:3))
#>       time
#> i      2021-01-04 2021-01-11 2021-01-18
#>   [1,]          0          0          0
#>   [2,]          0          0          0
#>   [3,]          0          0          0
#>   [4,]          0          0          0
#>   [5,]          0          0          0
#>   [6,]          0          0          0
birdflow_options(time_format = original_format)