OnlineStats

Online Algorithms for Statistics, Models, and Big Data Viz

Online algorithms are well suited for streaming data or when data is too large to hold in memory. OnlineStats processes observations one by one and all algorithms use O(1) memory.

Docs	Build	Test	Citation

Quickstart

import Pkg

Pkg.add("OnlineStats")

using OnlineStats

o = Series(Mean(), Variance(), P2Quantile(), Extrema())

fit!(o, 1.0)

fit!(o, randn(10^6))

Documentation

Contributing

• Trivial PRs such as fixing typos are very welcome!
  
• For nontrivial changes, you'll probably want to first discuss the changes via issue/email/slack with @joshday.

Authors

• Primary Author: Josh Day (@joshday)
• Significant early contributions from Tom Breloff (@tbreloff)
• Many algorithms developed under mentorship of Hua Zhou (@Hua-Zhou)

See also the list of contributors to OnlineStats.

License

MIT

Packages Using OnlineStats/OnlineStatsBase

See JuliaHub: