Welcome to Aeronet’s documentation!

Aim and scope

Aeronetlib is designed to make it easier for the deep learning researchers to handle the remote sensing data.

The main functionality of the library:

• handling of the geospatial rasters, including big images;
• random sampling of the image patches from a dataset for the model training;
• sequential sampling of the patches from an image for inference;
• transformation between raster mask and vector polygons in both directions.

• Requirements and installation
• Quickstart guide
  • Data preparation
  • Dataset exploration and model training
  • Inference
• Module reference
  • Raster
    • Band
    • Collection
    • GeoObject
  • Vector
    • Feature
    • FeatureCollection
  • Other
    • Transforms
    • Converters
    • Coords
    • Random dataset
    • IO
    • Dataset utils
    • Visualization

Indices and tables

• Index
• Module Index
• Search Page