Metadata-Version: 2.3
Name: cryocast
Version: 2026.10
Summary: CryoCast multimodal pipeline for predicting sea ice.
Author: Sophie Arana, Simon Driscoll, Isabel Fenton, Lydia France, Aoife Hughes, Maria Carolina Novitasari, Nilo Pedrazzini, Erin Quan, James Robinson, Shaerdan Shataer, Louisa van Zeeland
Author-email: Sophie Arana <arana.sophie@gmail.com>, Simon Driscoll <simondriscoll88@gmail.com>, Isabel Fenton <ifenton@turing.ac.uk>, Lydia France <85945427+LydiaFrance@users.noreply.github.com>, Aoife Hughes <aoife.1.hughes@gmail.com>, Maria Carolina Novitasari <124182131+marianovitasari20@users.noreply.github.com>, Nilo Pedrazzini <npedrazzini@turing.ac.uk>, Erin Quan <72717299+erinuclkwon@users.noreply.github.com>, James Robinson <james.em.robinson@gmail.com>, Shaerdan Shataer <Shaerdan@gmail.com>, Louisa van Zeeland <louisa-ai2@protonmail.com>
Requires-Dist: anemoi-datasets==0.5.47
Requires-Dist: anemoi-transform==0.5.1
Requires-Dist: anemoi-utils==0.5.14
Requires-Dist: argopy==1.5.0
Requires-Dist: cdsapi==0.7.7
Requires-Dist: earthkit-data==0.20.0
Requires-Dist: eccodeslib==2.47.3.21
Requires-Dist: filelock==4.2.4
Requires-Dist: haversine==2.9.0
Requires-Dist: hydra-core==1.3.7
Requires-Dist: imageio-ffmpeg==0.6.0
Requires-Dist: jaxtyping==0.3.11
Requires-Dist: lightning==2.6.6
Requires-Dist: matplotlib==3.11.2
Requires-Dist: netcdf4==1.7.5
Requires-Dist: numpy==2.4.6
Requires-Dist: omegaconf==2.3.1
Requires-Dist: optuna==5.0.0
Requires-Dist: pandas==2.3.3
Requires-Dist: pillow==13.0.0
Requires-Dist: pydantic==2.14.1
Requires-Dist: pyproj==3.7.2
Requires-Dist: pytorch-triton==3.5.1 ; (platform_machine == 'aarch64' and sys_platform == 'linux') or (platform_machine == 'x86_64' and sys_platform == 'linux')
Requires-Dist: pyyaml==6.0.3
Requires-Dist: scikit-learn==1.9.1
Requires-Dist: torch==2.14.1
Requires-Dist: torchmetrics==1.9.0
Requires-Dist: tqdm==4.70.1
Requires-Dist: typer==0.27.3
Requires-Dist: typing-extensions==4.16.0
Requires-Dist: wandb==0.31.0
Requires-Dist: xarray==2026.9.0
Requires-Dist: zarr==2.18.7
Requires-Python: >=3.11
Description-Content-Type: text/markdown

[![Tests](https://github.com/alan-turing-institute/cryocast/actions/workflows/test_code.yaml/badge.svg)](https://github.com/alan-turing-institute/cryocast/actions/workflows/test_code.yaml)
[![Docs](https://github.com/alan-turing-institute/cryocast/actions/workflows/build_docs.yml/badge.svg)](https://github.com/alan-turing-institute/cryocast/actions/workflows/build_docs.yml)
[![Code style](https://github.com/alan-turing-institute/cryocast/actions/workflows/code_style.yaml/badge.svg)](https://github.com/alan-turing-institute/cryocast/actions/workflows/code_style.yaml)
[![Licence: MIT](https://img.shields.io/badge/licence-MIT-green)](LICENSE)

CryoCast is a **multimodal machine-learning framework for sea-ice forecasting**. It combines satellite observations, Argo float sensor data, and ERA5 reanalysis fields to produce short-term Arctic and Antarctic sea-ice concentration forecasts.

Example forecasts are shown below.

**Arctic**

![Example CryoCast Arctic sea ice concentration forecast compared with observations](docs/src/assets/prediction-fullnorth-ddpm-v2026.07.png)

**Antarctic**

![Example CryoCast Antarctic sea ice concentration forecast compared with observations](docs/src/assets/prediction-fullsouth-ddpm-v2026.07.png)

The encode-process-decode architecture translates each input dataset into a shared latent space, allowing new data sources and model components to be added without changing the full pipeline.

## Key capabilities

- Multimodal data fusion across satellite observations, reanalysis fields, and in-situ sensor data.
- Extensible encode-process-decode design for adding new input sources and prediction targets.
- Multiple model configurations, including UNet, vision transformer, diffusion, and persistence-baseline approaches.

## Project context

CryoCast is developed at [The Alan Turing Institute](https://www.turing.ac.uk/) as a research system for Arctic and Antarctic sea-ice forecasting. The current codebase supports research, benchmarking, sensitivity experiments, and case-study analysis; it is not intended for operational forecasting, safety-critical decision making, or public warnings. See the [model card](docs/MODEL_CARD.md) for intended use and scope.

## Quick start

```bash
git clone git@github.com:alan-turing-institute/cryocast.git
cd cryocast
uv sync --managed-python
```

Create a local config in `cryocast/config/` (see [Configuration](https://alan-turing-institute.github.io/cryocast/user-guide/configuration/) for details):

```yaml
# cryocast/config/my.local.yaml
defaults:
  - base
  - _self_

base_path: /path/to/my/data
```

Then download datasets and train:

```bash
uv run cryocast datasets create --config-name my.local
uv run cryocast train --config-name my.local
```

Evaluate a checkpoint:

```bash
uv run cryocast evaluate --checkpoint /path/to/checkpoint.ckpt --config-name my.local
```

## Documentation

See the [project documentation](https://alan-turing-institute.github.io/cryocast/) for the full user guide and reference material.

- [Installation](https://alan-turing-institute.github.io/cryocast/user-guide/installation/) for prerequisites, `uv` setup, and HPC-specific steps
- [Configuration](https://alan-turing-institute.github.io/cryocast/user-guide/configuration/) for local config files, model overrides, and custom datasets
- [Commands](https://alan-turing-institute.github.io/cryocast/user-guide/commands/) for `datasets create`, `datasets inspect`, `train`, and `evaluate`
- [Add a model](https://alan-turing-institute.github.io/cryocast/how-to/add-a-model/) for tensor format and model architecture guidance

## Jupyter notebooks

The `notebooks/` folder contains demonstrator notebooks. Run them with:

```bash
uv run --group notebooks jupyter notebook
```

Start with `notebooks/demo_pipeline.ipynb` for a worked example of the full pipeline.

## Contributing

See [CONTRIBUTING.md](CONTRIBUTING.md) for the development workflow, coding conventions, and test instructions. For project questions, contact [SeaIce@turing.ac.uk](mailto:SeaIce@turing.ac.uk).

## License

CryoCast is released under the [MIT License](LICENSE).
