Images and Labels for SDSS Main Galaxy Sample (From Dey et al, 2022)

Data set used in the paper "Photometric redshifts from SDSS images with an interpretable deep capsule network" by Dey et al. (2022). The parent data set is from "Photometric redshifts from SDSS images using a convolutional neural network" by Pasquet et al. (2018). This data set adds quality cuts and merges additional labels and uses 16 bit floating point precision for storage. This data set is a good test bed for algorithms training on raw images to predict galaxy properties. Details of the data set can be found in the respective publications.