Usage

Loading a dataset can be as easy as:

from galaxies_datasets import datasets
import tensorflow_datasets as tfds

# Construct a tf.data.Dataset
ds = tfds.load("galaxy_zoo_challenge", split="train")

# Build your input pipeline
ds = ds.shuffle(1000).batch(128).prefetch(10).take(5)

In the example above:

from galaxies_datasets import datasets

registers the collection of galactic datasets with the tensorflow_datasets package making them available through its API. And that is it! …Almost.

For more details on tensorflow_datasets check out the documentation.

Some datasets require that you first manually download data. Check each dataset for instructions.

Datasets

Currently available datasets focus on galaxy morphology.

They include observational data from the Galaxy zoo project:

• galaxy_zoo_challenge

• galaxy_zoo2

• galaxy_zoo_decals

As well as mock galaxy images from the EAGLE simulation:

• eagle

Installation

You can install Galaxies Datasets via pip from PyPI:

$ pip install galaxies-datasets

Scripts

Galaxies Datasets provides some scripts to download and prepare data. The scripts are available through a command-line interface powered by Typer.

For example, to download images and data from the EAGLE simulation you could simply do:

galaxies_datasets eagle download USER SIMULATION

where USER is your username for the EAGLE public database and SIMULATION is the name of one of the EAGLE simulations.

For all available commands check the Command-line Interface reference, or run:

galaxies_datasets --help

The command-line interface also supports automatic completion in all operating systems, in all the shells (Bash, Zsh, Fish, PowerShell), so that you can just hit TAB and get the available options or subcommands.

To install automatic completion in bash run:

galaxies_datasets --install-completion bash

Citation

If you use this software, please cite it as below, in addition to any citation specific to the used datasets.

@software{lucas_bignone_2021_5521451,
    author       = {Lucas Bignone},
    title        = {Galaxies Datasets},
    month        = sep,
    year         = 2021,
    publisher    = {Zenodo},
    version      = {v0.1.1},
    doi          = {10.5281/zenodo.5521450},
    url          = {https://doi.org/10.5281/zenodo.5521450}
}

Contributing

Contributions are very welcome. To learn more, see the Contributor Guide.

License

Distributed under the terms of the MIT license, Galaxies Datasets is free and open source software.

Issues

If you encounter any problems, please file an issue along with a detailed description.

Disclaimer

This is a utility library that downloads and prepares datasets. We do not host or distribute these datasets, vouch for their quality or fairness, or claim that you have license to use the dataset. It is your responsibility to determine whether you have permission to use the dataset under the dataset’s license.

If you’re a dataset owner and wish to update any part of it (description, citation, etc.), or do not want your dataset to be included in this library, please get in touch through a GitHub issue. Thanks for your contribution to the ML community!

Credits

This project was generated from @cjolowicz’s Hypermodern Python Cookiecutter template.

Icons made by Freepik from www.flaticon.com