Implementation of the Aurora model for atmospheric forecasting.
The package currently includes the pretrained model and the fine-tuned version for high-resolution weather forecasting. We are working on the fine-tuned version for air pollution forecasting, which will be included in due time.
Please see the documentation for detailed instructions and more examples. You can also directly go to a full-fledged example that runs the model on ERA5.
Cite us as follows:
@misc{bodnar2024aurora,
title = {Aurora: A Foundation Model of the Atmosphere},
author = {Cristian Bodnar and Wessel P. Bruinsma and Ana Lucic and Megan Stanley and Johannes Brandstetter and Patrick Garvan and Maik Riechert and Jonathan Weyn and Haiyu Dong and Anna Vaughan and Jayesh K. Gupta and Kit Tambiratnam and Alex Archibald and Elizabeth Heider and Max Welling and Richard E. Turner and Paris Perdikaris},
year = {2024},
url = {https://arxiv.org/abs/2405.13063},
eprint = {2405.13063},
archivePrefix = {arXiv},
primaryClass = {physics.ao-ph},
}
Contents:
- What is Aurora?
- Getting Started
- Contributing
- License
- Security
- Responsible AI Transparency Documentation
- Trademarks
- FAQ
Aurora is a machine learning model that can predict atmospheric variables, such as temperature. It is a foundation model, which means that it was first generally trained on a lot of data, and then can adapted to specialised atmospheric forecasting tasks with relatively little data. We provide three such specialised versions: one for medium-resolution weather prediction, one for high-resolution weather prediction, and one for air pollution prediction.
Install with pip:
pip install microsoft-auroraRun the pretrained small model on random data:
from datetime import datetime
import torch
from aurora import AuroraSmall, Batch, Metadata
model = AuroraSmall()
model.load_checkpoint("microsoft/aurora", "aurora-0.25-small-pretrained.ckpt")
batch = Batch(
surf_vars={k: torch.randn(1, 2, 17, 32) for k in ("2t", "10u", "10v", "msl")},
static_vars={k: torch.randn(17, 32) for k in ("lsm", "z", "slt")},
atmos_vars={k: torch.randn(1, 2, 4, 17, 32) for k in ("z", "u", "v", "t", "q")},
metadata=Metadata(
lat=torch.linspace(90, -90, 17),
lon=torch.linspace(0, 360, 32 + 1)[:-1],
time=(datetime(2020, 6, 1, 12, 0),),
atmos_levels=(100, 250, 500, 850),
),
)
prediction = model.forward(batch)
print(prediction.surf_vars["2t"])Note that this will incur a 500 MB download.
Please read the documentation for more detailed instructions and for which models are available.
See CONTRIBUTING.md.
See LICENSE.txt.
See SECURITY.md.
An AI system includes not only the technology, but also the people who will use it, the people who will be affected by it, and the environment in which it is deployed. Creating a system that is fit for its intended purpose requires an understanding of how the technology works, its capabilities and limitations, and how to achieve the best performance. Microsoft has a broad effort to put our AI principles into practice. To find out more, see Responsible AI principles from Microsoft.
Our goal in publishing this code is (1) to facilitate reproducibility of our paper and (2) to support and accelerate further research into foundation model for atmospheric forecasting. This code has not been developed nor tested for non-academic purposes and hence should not be used as such.
Although Aurora was trained to accurately predict future weather and air pollution, Aurora is based on neural networks, which means that there are no strict guarantees that predicts will always be accurate. Altering the inputs, providing a sample that was not in the training set, or even providing a sample that was in the training set but is simply unlucky may result in arbitrarily poor predictions. In addition, even though Aurora was trained on a wide variety of data sets, it is possible that Aurora inherits biases present in any one of those data sets. A forecasting system like Aurora is only one piece of the puzzle in a weather prediction pipeline, and its outputs are not meant to be directly used by people or business to plan their operations. A series of additional verification tests are needed before it could become operationally useful.
The models included in the code have been trained on a variety of publicly available data. A description of all data, including download links, can be found in Supplementary C of the paper.
All versions of Aurora were extensively evaluated by evaluating predictions on data not seen during training. These evaluations not only compare measures of accuracy, such as the root mean square error and anomaly correlation coefficient, but also look at the behaviour in extreme situations, like extreme heat and cold, and rare events, like Storm Ciarán in 2023. These evaluations are the main topic of the paper.
Note: The documentation included in this file is for informational purposes only and is not intended to supersede the applicable license terms.
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.
First, install the repository in editable mode and setup pre-commit:
make installTo run the tests and print coverage, run
make testYou can then explore the coverage in the browser by opening htmlcov/index.html.
To locally build the documentation, run
make docsTo locally view the documentation, open docs/_build/index.html in your browser.
The package currently includes the pretrained model and the fine-tuned version for high-resolution weather forecasting. We are working on the fine-tuned version for air pollution forecasting, which will be included in due time.