PSCAE-SN: Physics-Aware Spiking RFI Detection for Simulated MeerKAT Data.
This archive contains version 1.0.0 of PSCAE-SN, a physics-aware spiking convolutional autoencoder for detecting radio-frequency interference in simulated MeerKAT data. It includes the executed Jupyter notebook used for model training, evaluation, and figure generation, together with citation metadata, documentation, and the MIT License. The notebook reproduces five stochastic training runs and the reported pixel-level detection metrics using the publicly available TABASCAL SNN-NLN dataset. Trained model weights are not included; users must retrain the model by executing the notebook.