ForestsKing/D3R
PyTorch implementation of "Drift doesn't Matter: Dynamic Decomposition with Dffusion Reconstruction for Unstable Multivariate Time Series Anomaly Detection" (NeurIPS 2023)
This project helps operations engineers and data analysts detect unusual activity in complex systems where conditions often change. It takes in streams of sensor data or other multivariate time series and outputs signals indicating potential anomalies, even when the system itself is undergoing normal 'drift' or shifts in behavior. This is ideal for monitoring industrial control systems, network performance, or environmental sensors.
No commits in the last 6 months.
Use this if you need to reliably find anomalies in real-time data from systems that naturally evolve or experience changing operational conditions, without getting flooded by false alarms.
Not ideal if your time series data is perfectly stable with no expected shifts over time, or if you only have a single data stream rather than multiple correlated metrics.
Stars
86
Forks
12
Language
Python
License
—
Category
Last pushed
Jul 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/ForestsKing/D3R"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
yang-song/score_sde_pytorch
PyTorch implementation for Score-Based Generative Modeling through Stochastic Differential...
ermongroup/ncsnv2
The official PyTorch implementation for NCSNv2 (NeurIPS 2020)
yang-song/score_sde
Official code for Score-Based Generative Modeling through Stochastic Differential Equations...
amazon-science/unconditional-time-series-diffusion
Official PyTorch implementation of TSDiff models presented in the NeurIPS 2023 paper "Predict,...
AI4HealthUOL/SSSD-ECG
Repository for the paper: 'Diffusion-based Conditional ECG Generation with Structured State Space Models'