Methods & Tools For Time Series Data Science Problems With InfluxDB
summary
Shows a working pipeline that uses the Telegraf execd processor to micro‑batch sensor lines, call a Python script using statsmodels implementing double exponential smoothing (Holt’s method), write forecasts and residuals back to InfluxDB, and flag anomalies (demo finds six anomalous points). Also demonstrates server‑side analytics with Flux including MAD anomaly detection and live use from Jupyter and Zeppelin integration, with code and Telegraph configs published on the project GitHub and runnable against a simulator.
AI-written summary · about
chapters (8)
- 0:00Introduction to influxDB
- 2:00Forecasting with Telegraf
- 6:18Implementing forecasting
- 10:53Analytics and anomalies
- 16:13Data processing with flux
- 21:12Notebook integrations
- 23:05Data science libraries
- 25:46Final demo and resources
related talks
28mTime series data science problems with InfluxDB
44mMethods and Tools for Time Series Data Science Problems
47mBack to the Future of Time Series Data
47mBack to the Future of Time Series Data
49mThinking in time series, the new solution to your pro...
12mTime to awesome: InfluxDB and the AWS environment
23mTime series database: Should I use one?
41mMastering real-time anomaly detection with open source tools