r.learn.predict.worker - GRASS manual

NAME

r.learn.predict.worker - Applies a classification model to a region. It is called in parallel by r.learn.parallel.predict.

KEYWORDS

raster, classification, regression, machine learning, scikit-learn, prediction, parallel

SYNOPSIS

r.learn.predict.worker

r.learn.predict.worker --help

r.learn.predict.worker [-pz] group=name load_model=name output=name [chunksize=integer] mapset=name area=name where=sql_query [nsres=value] [ewres=value] [--overwrite] [--help] [--verbose] [--quiet] [--ui]

Flags:

-p: Output class membership probabilities; A raster layer is created for each class. For the case of a binary classification, only the positive (maximum) class is output
-z: Only predict class probabilities
--overwrite: Allow output files to overwrite existing files
--help: Print usage summary
--verbose: Verbose module output
--quiet: Quiet module output
--ui: Force launching GUI dialog

Parameters:

group=name [required]: Group of raster layers used for prediction; GRASS imagery group of raster maps representing feature variables to be used in the machine learning model
load_model=name [required]: Load model from file; File representing pickled scikit-learn estimator model
output=name [required]: Output Map; Raster layer name to store result from classification or regression model. The name will also used as a perfix if class probabilities or intermediate of cross-validation results are ordered as maps.
chunksize=integer: Number of pixels to pass to the prediction method; Number of pixels to pass to the prediction method. GRASS GIS reads raster by-row so chunksize is rounded down based on the number of columns; Default: 100000
mapset=name [required]: Name of mapset to create
area=name [required]: Input vector map to set the region with a where condition; Or data source for direct OGR access
where=sql_query [required]: Where condition to set the region to the area; Example: income < 1000 and population >= 10000
nsres=value: North-south 2D grid resolution
ewres=value: East-west 2D grid resolution