Benjamin Leutner
Benjamin Leutner
Not entirely sure whether your bugtracker is the correct one for this issue (happens in eclipse shellwax), so feel free to send me elsewhere, if not. Failed to parse: $((...
# Function transition Transition all RStoolbox spatial functions to use only terra and sf, while maintaining backward compatibility to sp and raster. See [switch_to_terra](https://github.com/bleutner/RStoolbox/tree/switch_to_terra) branch ⚠️ table not yet completely...
note to self: * everything with doParallel backend (.parXapply) is fine * only .paraRasterFun apparently doesn't cooperate (Windows only), e.g. predict in superClass
Feature request: Wallis filter for local contrast enhancement Example: http://www.mdpi.com/2072-4292/8/3/178/htm
Feature request: full unmixing workflow Relevant for `sam()` and `mesma()`
memory consumption can be too high due to single layer processing. Todo: * add manual canProcessInMemory checks for sum of all output layers. * add ellipsis filename support
Resources: * An illumination correction algorithm on Landsat-TM data http://ieeexplore.ieee.org/document/5653492/ * http://www.arcgis.com/home/item.html?id=fceca6e14fcc4d25953958a52f3eba5b * "Improved forest change detection with terrain illumination corrected Landsat images" https://www.sciencedirect.com/science/article/pii/S0034425713001673?via%3Dihub#f0025
When specifying the holdout samples `indexOut` for cross-validation myself, the `index` values reported in the train object are wrong. They are are created as if `indexOut` was not provided at...