m.neural_network - Toolset for creating training data and training a neural network

The m.neural_network toolset consists of following modules:

• m.neural_network.tindex: create tile index for data preparation as first step for the process of creating a neural network
• m.neural_network.preparedata_part1: prepare training and/or apply data as first step for the process of creating a neural network.
  • m.neural_network.preparedata_part1.worker_nullcells: Worker module for m.neural_network.preparedata_part1 to check null cells
  • m.neural_network.preparedata_part1.worker_export: Worker module for m.neural_network.preparedata_part1 to export data
• m.neural_network.preparedata_part2: prepare training and/or apply data for use in model training and application
  • m.neural_network.preparedata_part2.worker_label: Worker module for m.neural_network.preparedata_part2 to check and rasterize label data
• m.neural_network.train: training of a semantic segmentation model with smp libraries
• m.neural_network.test: calculation of statistics for quality assessment
• m.neural_network.apply: application of a trained model to new data
• m.neural_network.postprocessing.patch: patches the tiles (GeoTIFFs) which results from neural network inference
• m.neural_network.postprocessing.vectorize: vectorizes the classification raster output and clean results (remove small areas, if set straighten lines)
• m.neural_network.postprocessing.snapref: snaps classification vector with reference data.

Additional required GRASS GIS addons

• v.out.geojson from mundialis