Skip to main content

Spatiotemporal phenology research with interpretable models

Project description

Documentation Status RSD DOI PyPI version

For detailed information and instruction, please refer to the documentation

Springtime

Springtime is both a project and a python packaged aimed at streamlining workflows for doing machine learning with phenological datasets.

Phenology is the scientific discipline in which we study the lifecycle of plants and animals. A common objective is to develop (Machine Learning) models that can explain or predict the occurrence of phenological events, such as the blooming of plants. Since there is a variety of data sources and existing tools to retrieve and analyse phenology data, it is easy to get lost and disorganized.

At the heart of springtime is a data representation following the scikit-learn standard structure. The springtime python package implements (down)loaders for various datasets that are able to convert the data to this same structure. Data loading specifications can be exported to yaml recipes for easy sharing.

The documentation has an extensive user guide that shows how each of the data loaders convert from the raw to the standardized data format. It also includes examples of using various (combinations of) models.

The data structure proposed here is still not ideal, and should rather be seen as a first step in standardizing workflows in phenological modelling. We hope it will serve as a basis for discussion and further developments.

Example task

Predict the day of first bloom of the common lilac given indirect observations (e.g. satellite data) and/or other indicators (e.g. sunshine and temperature).

illustration_example_use_case

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

springtime-0.2.2.tar.gz (37.1 kB view hashes)

Uploaded Source

Built Distribution

springtime-0.2.2-py3-none-any.whl (47.5 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page