Skip to main content

Tools for processing and evaluating anemoi based EAGLE ML models.

Project description

eagle-tools

Tools for processing and evaluating anemoi based EAGLE ML models

⚠️ Disclaimer ⚠️

This package is pip-installable, but it is more in the form of research code rather than well-documented and tested software. There are likely better and more efficient ways to accomplish the main functionality of this package, but this gets the job done.

Installation

Since some dependencies are only available on conda, it's recommended to create a conda environment for all dependencies. Note that this package is not (yet) available on conda, but it can still be installed via pip.

conda env create -n eagle
conda install -c conda-forge ufs2arco
pip install anemoi-datasets anemoi-graphs anemoi-models anemoi-training anemoi-inference anemoi-utils anemoi-transform
pip install git+https://github.com/timothyas/xmovie.git@feature/gif-scale
pip install eagle-tools

Usage

This provides the following functionality. Note that each command uses a configuration yaml, and documentation of the yaml contents can be found by running eagle-tools <command> --help. For example, one can run eagle-tools inference --help to get documentation.

Inference

Run anemoi-inference over many initial conditions

eagle-tools inference config.yaml

Averaged Error Metrics

Compute Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE), preserving the initial condition dimension (t0).

eagle-tools metrics config.yaml

Spatial Error Metrics

Compute the spatial distribution of RMSE and MAE for each lead time. By default, these are averaged over all initial conditions used.

eagle-tools spatial config.yaml

Power Spectra

Compute the power spectra, averaged of initial conditions.

eagle-tools spectra config.yaml

Visualize Predictions Compared to Targets

Make figures or movies, showing the targets and predictions. Note that the argument end_date has different meanings for each. For figures, end_date is the date plotted, whereas for movies, all timestamps between start_date and end_date get shown in the movie.

eagle-tools figures config.yaml
eagle-tools movies config.yaml

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

eagle_tools-0.0.0.tar.gz (23.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

eagle_tools-0.0.0-py3-none-any.whl (30.2 kB view details)

Uploaded Python 3

File details

Details for the file eagle_tools-0.0.0.tar.gz.

File metadata

  • Download URL: eagle_tools-0.0.0.tar.gz
  • Upload date:
  • Size: 23.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for eagle_tools-0.0.0.tar.gz
Algorithm Hash digest
SHA256 e3ceac42b5c210e38384d787373edd4b5449b15cf6da56cf771a3392c24a3db4
MD5 03bd04cb7a78134c989d63cce36c733c
BLAKE2b-256 36281ee0d2ff483a6fbf699aafc02e2f3450409f863191112239bd613fe28b49

See more details on using hashes here.

Provenance

The following attestation bundles were made for eagle_tools-0.0.0.tar.gz:

Publisher: publish.yaml on NOAA-PSL/eagle-tools

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

File details

Details for the file eagle_tools-0.0.0-py3-none-any.whl.

File metadata

  • Download URL: eagle_tools-0.0.0-py3-none-any.whl
  • Upload date:
  • Size: 30.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/6.1.0 CPython/3.12.9

File hashes

Hashes for eagle_tools-0.0.0-py3-none-any.whl
Algorithm Hash digest
SHA256 8cb0215087a49ed8cf01b4953ae9153c25893f16f646546af65d18de5614fc89
MD5 d5292ae86b3fd6a7289cf4b37bfd7d88
BLAKE2b-256 4b00433709ff1dd2fa08a93df78b3e57b977f1f34f8237ba379d4c2b4684279f

See more details on using hashes here.

Provenance

The following attestation bundles were made for eagle_tools-0.0.0-py3-none-any.whl:

Publisher: publish.yaml on NOAA-PSL/eagle-tools

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Supported by

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