About Ensemble Learning Models

Ensemble Learning Models (elm ) is a set of tools for parallel ensemble unsupervised and supervised machine learning, with a focus on data structures useful in climate science and satellite imagery.

elm Capabilities

• Ensemble learning
• Large scale prediction
• Genetic algorithms
• Common preprocessing operations for satellite imagery and climate data

These capabilities are best shown in the

• Elm introduction
• Elm clustering introduction
• Other elm examples
• Use cases

elm wraps together the following Python packages:

• dask-distributed: elm uses dask-distributed for parallelism over ensemble fitting and prediction
• scikit-learn : elm can use unsupervised and supervised models, preprocessors, scoring functions, and postprocessors from scikit-learn or any estimator that follows the scikit-learn initialize / fit / predict estimator interface.
• xarray : elm wraps xarray data structures for n-dimensional arrays, such as 3-dimensional weather cubes, and for collections of 2-D rasters, such as a LANDSAT sample

elm is a Work in Progress

elm is immature and largely for experimental use.

The developers do not promise backwards compatibility with future versions.

Next steps

• Install elm
• Try the example notebooks