This page contains the accompanying deep learning models, dataset and code for the paper on MargNet, titled "Photometric identification of compact galaxies, stars and quasars using multiple neural networks". Deep Learning Models: MargNet is a deep learning-based classifier for identifying stars, quasars and compact galaxies using photometric parameters and images from the Sloan Digital Sky Survey. MargNet consists of a combination of Convolutional Neural Network (CNN) and Artificial Neural Network (ANN) architectures. The deep learning Keras model for each experiment was saved as an h5 file after training. All saved models (organised by different experiments, as described in the paper) are available in SavedModels.zip. Dataset: Our dataset consists of 240,000 compact objects and an additional 150,000 faint objects consisting of an equal number of stars, galaxies and quasars. This data is available as NumPy arrays and CSV files, as described below: SDSS ObjID of each object (objlist.npy) SDSS 5-band images of each object cropped to 32*32 pixels (X.npy) The set of 24 photometric features for each object (dnnx.npy) The classification label for each object (y.npy) SDSS spreadsheet containing all the features from dnnx, labels from y, ObjIDs from objlist and a couple of more SDSS specific parameters (photofeatures.csv) The complete dataset (organised by different experiments, as described in the paper) is available in Dataset.zip. (Note: objlist, X, dnnx and y are in the same order. So, objlist[0], X[0], dnnx[0] and y[0] correspond to the same object.) Code: All our code was written in Python in the form of Jupyter Notebooks. A copy of our code has also been made available on GitHub, but not all files could be included on GitHub due to the storage limit. So a complete copy of the repository has also been mirrored here on Zenodo and is contained in MargNet_RepositoryMirror.tgz