Real-bogus scores for active anomaly detection

<p>Data description for <a href="https://arxiv.org/abs/2409.10256">Semenikhin et al., 2024</a></p> <p>The dataset consists of the following files:</p> <p><strong>feature_snad4_r_100.dat</strong> contains light curve feature data for objects, where each object is represented by 54 feature values. These values are encoded as little-endian single-precision IEEE-754 floating-point numbers (32-bit floats). Feature names are listed in the plain text file <strong>feature_snad4_r_100.name</strong>, with one name per line.<br><strong>sid_snad4_r_100.dat</strong> contains ZTF DR object identifiers, encoded as little-endian 64-bit unsigned integers.</p> <p><strong>exp_feature_snad4_r_100.dat</strong> contains the same features as <strong>feature_snad4_r_100.dat</strong>, but with an additional column representing the real-bogus classifier prediction. Each object in this file corresponds to 55 features: the original 54 features plus 1 additional feature. Feature names for this file are provided in <strong>exp_feature_snad4_r_100.name</strong>.</p> <p>The files <strong>sid_snad4_r_100.dat</strong>, <strong>feature_snad4_r_100.dat</strong>, and <strong>exp_feature_snad4_r_100.dat</strong> share the same object order.</p> <p><br>Below is a sample Python script for accessing the data using NumPy:</p> <p><code>import numpy as np</code></p> <p><code># Load object IDs</code><br><code>oid &nbsp; &nbsp; = &nbsp;np.memmap('sid_snad4_r_100.dat', mode='c', dtype=np.uint64)</code></p> <p><code># Load features and reshape</code><br><code>feature = np.memmap('feature_snad4_r_100.dat', mode='c', dtype=np.float32).reshape(oid.shape[0], -1)</code></p> <p><code># Print dataset information</code><br><code>print(f'Number of objects: {len(oid)}')</code><br><code>print(f'Features shape: {feature.shape}')</code></p>