WenjieDu/PyPOTS

A Python toolkit/library for reality-centric machine/deep learning and data mining on partially-observed time series, including SOTA neural network models for scientific analysis tasks of imputation/classification/clustering/forecasting/anomaly detection/cleaning on incomplete industrial (irregularly-sampled) multivariate TS with NaN missing values

73
/ 100
Verified

This project helps scientists, operations engineers, and other practitioners analyze real-world time series data that often has gaps or missing values. It takes your raw, incomplete time series data and can fill in the missing parts, predict future trends, classify patterns, group similar sequences, or detect unusual events. The result is clean, actionable insights from your time-based measurements.

1,965 stars. Used by 1 other package. Actively maintained with 16 commits in the last 30 days. Available on PyPI.

Use this if you need to perform advanced analysis like forecasting or anomaly detection on industrial or scientific time series data that frequently contains missing observations.

Not ideal if your time series data is always complete and perfectly regularly sampled, as it specifically addresses challenges with partially-observed data.

industrial-monitoring scientific-analysis predictive-maintenance sensor-data-processing time-series-forecasting
Maintenance 17 / 25
Adoption 11 / 25
Maturity 25 / 25
Community 20 / 25

How are scores calculated?

Stars

1,965

Forks

180

Language

Python

License

BSD-3-Clause

Last pushed

Mar 12, 2026

Commits (30d)

16

Dependencies

16

Reverse dependents

1

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/WenjieDu/PyPOTS"

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