We propose an anomaly detection method based on modal representation and a noise-robust sparse sensor position optimization method. We focus on the detection of anomalies in global sea surface ...

🚀 The feature, motivation and pitch Hi, IMHO: There are several problems with torch.auto_grad.detect_anomaly(): for NaN detection: It raises errors inside pytorch internals which can't be debugged by ...

Anomaly detection is the cornerstone of the health management of much large industrial mechanical equipment. Most machinery anomaly detection methods try to find a variable threshold to indicate ...

Deep residual autoencoder for reconstructing and analyzing spectral data using PyTorch. Includes composite loss, UMAP visualization, and spectral diagnostics. Built for unsupervised learning on ...