Package structure overview#
This section documents the general principles underpinning the organisation of the Eradiate Python package.
Main entry points#
- Core support [
The top-level module contains basic support functions required to do almost anything with Eradiate.
- Scene generation [
This package exposes the scene generation components. It includes convenient interfaces to create objects representing the geometry, optical properties, illumination and measures in the scene on which you’ll run your radiative transfer simulation.
- Experiments [
In this package, you will find interfaces to trigger simulation runs, including pre- and post-processing operations.
- Kernel [
This module provides functionality related with Eradiate’s radiometric kernel Mitsuba.
Data handling and visualisation#
- Data handling [
This package serves data shipped with Eradiate.
- Post-processing pipeline definitions [
This package provides a post-processing pipeline framework used to convert raw sensor results yielded by kernel sensors to quantities of interest for Earth observation applications (e.g. reflectance). The data is stored as xarray labelled arrays (Dataset).
- xarray utility functions [
Various support components taking advantage of the xarray library.
- Plotting [
This module defines optional utility functions to create Matplotlib-based visualisations of Eradiate’s input and output data.
- Radiative properties [
This package provides abstractions used to define radiative properties used to create scenes.
- Thermosphysical properties [
This package provides abstractions used to define thermophysical properties of scene objects. The output of its components are generally used as input of components responsible for radiative property computation.
- Numerical constants [
Various numerical constants used throughout the code base.
Other support components#
- Unit support [
Various utility functions and data variables used to safely handle unit conversions in Eradiate.
- Math support [
For the cases where Eradiate’s math dependencies and kernel are not enough, additional mathematical tools are provided.
- Random number generation [
Support components for fine control of random number generation.
- Exceptions [
Exception and warning definitions.
- Class writing facilities [
- Miscellaneous [
Other support components which don’t fit in any of the aforementioned classification entries.