EQTPartners/TSDE

TSDE is a novel SSL framework for TSRL, the first of its kind, effectively harnessing a diffusion process, conditioned on an innovative dual-orthogonal Transformer encoder architecture with a crossover mechanism, and employing a unique IIF mask strategy (KDD 2024, main research track).

40
/ 100
Emerging

This project provides advanced self-supervised learning for time series data, enabling accurate predictions and insights even with incomplete information. It takes raw time series data (like sensor readings, stock prices, or patient vitals) and can fill in missing gaps, forecast future values, detect anomalies, or group similar patterns. Data scientists and machine learning engineers working with complex sequential datasets can use this for enhanced time series analysis.

No commits in the last 6 months.

Use this if you need to perform imputation, interpolation, forecasting, anomaly detection, classification, or clustering on time series data with state-of-the-art accuracy, especially when dealing with missing data points.

Not ideal if you are a business user looking for a no-code solution or someone without a strong background in machine learning and Python programming.

time-series-analysis predictive-modeling data-imputation anomaly-detection machine-learning-engineering
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 17 / 25

How are scores calculated?

Stars

32

Forks

8

Language

Jupyter Notebook

License

MIT

Last pushed

Aug 13, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/embeddings/EQTPartners/TSDE"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.