search

zalandoresearch / pytorch-ts

PyTorch based Probabilistic Time Series forecasting framework based on GluonTS backend
MIT License
1.25k stars 192 forks source link

Anomaly detection #36

Open rruizdeaustri opened 3 years ago

rruizdeaustri commented 3 years ago

Hi, would it be possible to use the code for anomaly detection ? I'm interested in applying the conditional normalizing flows model for this in the framework of detecting gravitational waves and compare with the performance of other models like VAE + LSTM, GANs, etc ...

Thanks !!

kashif commented 3 years ago

Thanks @rruizdeaustri for your interest. Surprisingly (Nalisnick et al., 2019) showed that likelihood-based models like flows etc. assign higher likelihoods to Out-of-Distribution (OOD) data, so I would say this model would not work well for OOD tasks. However, we have something in mind which would work so if you can email me we can take this discussion off-line, Thanks!