linkedin/greykite

Health v2 · maintenance only

Adoption not part of the score

1855 stars · 107 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Greykite is a Python time series forecasting library from LinkedIn built around its Silverkite algorithm, with support for trend, seasonality, changepoints, holidays, and autoregression. It includes an anomaly detection extension, exploratory plotting, grid search, and benchmarking tooling.

Use cases

• forecast business metrics time series
• detect anomalies in metric monitoring
• model holiday and event effects in forecasts
• grid search forecast model parameters
• explore seasonality and changepoints in time series
• benchmark forecasting algorithms

When to choose

• you need interpretable forecasts with trend, seasonality, changepoints, and holiday effects
• you want an end-to-end framework covering EDA, preprocessing, grid search, and evaluation
• you need tuned confidence intervals for metric anomaly alerting

When to avoid

• you need deep-learning forecasting models like N-BEATS or Transformers as the primary approach
• you want real-time streaming anomaly detection rather than batch forecasting
• you need a maintained library with frequent releases and broad community support

Facets

library · maturity active

machine-learning data-science benchmarking data-visualization data-science time-series analytics machine-learning python cross-platform time-series-forecasting anomaly-detection silverkite seasonality changepoint-detection holiday-effects

Member repositories

For agents

markdown · JSON · MCP: product_card(name="linkedin/greykite")