NinoLearn

NinoLearn is a research framework for the application of machine learning (ML) methods for the prediction of the El Nino-Southern Oscillation (ENSO).

It contains methods for downloading relevant data from their sources, reading raw data, postprocessing it and then access the postprocessed data in an easy way.

Moreover, it contains models for the ENSO forecasting:

· Deep Ensemble Model (DEM)

· Encoder-Decoder Ensemble Model

Installation for own development

1. Fork the repository.
2. Clone the repository to your local machine.

git clone https://github.com/Your_Username/ninolearn

3. Make a conda environment from the .yml file.

conda env create -f ninolearn.yml

4. Activate the environment.

conda activate ninolearn

5. Add ninolearn to the conda environment in 'development mode'.

conda develop /path/to/ninolearn

6. Fill out the ninolearn/private_template.py file with the required pathes and save a copy as private.py. The private.py will not be pushed to your remote repository because it contain sensitive information as well as pathes that are specific to your machine.

Now you should be ready to use ninolearn. For the beginning you can try to run the Jupyter Notebook tutorials which are currently located in docs-sphinx/source/jupyter_notebook_tutorials/.

Folder structure

In the folder ninolearn the actual ninolearn code is located. The research folder contains pervious research that was done with ninolearn. Hence, if you want to do your own research with ninolearn, make a new directory in the research folder in which you can start to do your own stuff.

The folders docs and docs-sphinx contain the documentation of ninolearn. Currently the documentation is somewhat outdated.