Galaxy images labelled by morphology (shape). Aimed at ML debugging and teaching.

Contains 10,000 images of galaxies (either 3x64x64 or 3x224x224), confidently labelled by Galaxy Zoo volunteers as belonging to one of four morphology classes.

git clone https://github.com/mwalmsley/galaxy_mnist
pip install -e galaxy_mnist

The only dependencies are pandas, scikit-learn, and h5py (for .hdf5 support). (py)torch and torchvision are required but not specified as a dependency, because you likely already have it and may require a very specific version (e.g. from conda, AWS-optimised, etc).

torch must be at least 1.9.1. Currently tested on torch 1.10.2 and torchvision 0.11.3.

Simply use as with MNIST:

from galaxy_mnist import GalaxyMNIST

# 64 pixel images
dataset = GalaxyMNIST(
    root='/some/download/folder',
    download=True,
    train=True  # by default, or set False for test set
)

Access the images and labels - in a fixed "canonical" 80/20 train/test division - like so:

images, labels = dataset.data, dataset.targets

You can also divide the data according to your own to your own preferences with load_custom_data:

(custom_train_images, custom_train_labels), (custom_test_images, custom_test_labels) = dataset.load_custom_data(test_size=0.8, stratify=True)

When you request either dataset[0] or datset.data, dataset.targets, you will recieve data according to whichever method you last called - either init or load_custom_data.

(New: For 224 pixel images, import and use GalaxyMNISTHighrez instead).

from galaxy_mnist import GalaxyMNISTHighrez

# 224 pixel images
dataset = GalaxyMNISTHighrez(...

See load_in_pytorch.py for a working example. For debugging help, see the FAQ section below.

Like MNIST, the data type is uint8. You may wish to convert to float before doing any math.

GalaxyMNIST has four classes: smooth and round, smooth and cigar-shaped, edge-on-disk, and unbarred spiral (you can retrieve this as a list with GalaxyMNIST.classes).

The galaxies are selected from Galaxy Zoo DECaLS Campaign A (GZD-A), which classified images taken by DECaLS and released in DR1 and 2. The images are as shown to volunteers on Galaxy Zoo, except for a 75% crop followed by a resize to 64x64 pixels.

At least 17 people must have been asked the necessary questions, and at least half of them must have answered with the given class. The class labels are therefore much more confident than from, for example, simply labelling with the most common answer to some question.

The classes are balanced exactly equally across the whole dataset (2500 galaxies per class), but only approximately equally (by random sampling) in the canonical train/test split. For a split with exactly equal classes on both sides, use load_custom_data with stratify=True.

You can see the exact choices made to select the galaxies and labels under the reproduce folder. This includes the notebook exploring and selecting choices for pruning the decision tree, and the script for saving the final dataset(s).

If you use this dataset, please cite Galaxy Zoo DECaLS, the data release paper from which the labels are drawn. Please also acknowledge the DECaLS survey (see the linked paper for an example).

You can find the original volunteer votes (and images) on Zenodo here.

Q: Error: "No such file or directory: ...train-images-idx3-ubyte

A: Update pytorch/torchvision (to at least 1.9.1). GalaxyMNIST is a subclass of torchvision.datasets.mnist and assumes this file is up-to-date.

Q: Error: galaxy-mnist is not a valid editable requirement. It should either be a path to a local project or a VCS URL (beginning with svn+, git+, hg+, or bzr+).

A: Check you are pointing pip install -e galaxy_mnist at the folder where you cloned the repo. For example, the error above is caused by mistakenly doing pip install -e galaxy-mnist with a - instead of a _.

If you're still having trouble, please raise an issue on GitHub or reach out directly to michael.walmsley@manchester.ac.uk.