User Guide

If you are new to xeofs, this is the place to start. Here, you will learn why xeofs was created, what it can do for you, and how to get started.

Advantages of using xeofs:

• Multi-Dimensional & Labeled Data: Designed for xarray objects, xeofs applies dimensionality reduction to multi-dimensional data while maintaining data labels. It works with both DataArray and Dataset objects, providing output that matches the type of input, whether single or a list of xr.DataArray or xr.Dataset.

• Dask-Integrated: Supports large datasets via dask xarray objects.

• Efficient: Ensures computational efficiency, especially with large datasets, through randomized SVD.

• Extensive Methods: Offers various dimensionality reduction techniques. For more details, see the API reference.

• Handling Missing Values: Can manage common cases where some features are masked out by NaN values, such as masked ocean or land cells.

• Bootstrapping: Provides a user-friendly interface for model evaluation using bootstrapping.

• Modular: Allows users to implement and incorporate new dimensionality reduction methods.

If you’re eager to see it in action, check out the basic example to get started quickly. For more comprehensive demonstrations, explore our example gallery.

Finally, if you’re using xeofs in your academic work, please consider citing the software. For more information, see how to cite.