R/quantity_allocation_disagreement_lite.R
quantity_allocation_disagreement_lite.Rd
`config_qad_lite` handles pre-processing, preparing the reference and comparison raster files, and saves them as an RDS file. This RDS is then used by `quantity_allocation_disagreement_lite` to compute metrics efficiently.
quantity_allocation_disagreement_lite(
config_qad_lite_file,
new_dirs_path = NULL,
use_configuration = NULL,
use_distance = NULL
)
the raster with ground truth data. For all metrics expect RMSE these values are reclassified with values > 1 becoming 1, values < 1
(Default = `NULL`) Specify a new root directory to update input and output file paths in the `config_qad_lite_file`. This helps adapt the file paths when running the configuration on a different workstation with a different root path, without needing to recreate the entire `config_qad_lite_file`. If `new_dirs_path` is `NULL` (default), the original paths in `config_qad_lite_file` are used. When specified, `new_dirs_path` should point to the top-level folder containing the input files and output folder. The folder structure under this top-level directory must match the structure in the original `config_qad_lite_file` If no match is found, the original input file paths and output folder remain unchanged.
Boolean if you want to use configuration disagreement for comparing model runs. NOTE: overrides boolean provided in the quantity, allocation, and disagreement config file.
Boolean if you want to compare distance between simulations and observations. NOTE: overrides boolean provided in the quantity, allocation, and disagreement config file.
A data frame with spatial configuration metrics. Particularly quantity, allocation, and total disagreement, omission and commission, and directional disagreement where directional disagreement.
Separating pre-processing from metric calculation minimizes memory usage in `quantity_allocation_disagreement_lite`, making it suitable for large extent, fine-resolution rasters.
Uses quantity and allocation disagreement metrics by Pontius and Millones (2014) and omission and commission errors of comparing a modeled raster data set to a reference raster data set.