ParallelKMeans.jl

ParallelKMeans v1.1.0

Diff since v1.0.1

Merged pull requests:

• Fixed error in benchmark plot & CSV (#116) (@PyDataBlog)
• One small typo in readme (#117) (@anderson15)
• Extend compat: StatsBase = "0.32, 0.33, 0.34" (#119) (@ablaom)
• Breaking change semantic version (#120) (@PyDataBlog)

ParallelKMeans v1.0.1

Diff since v1.0.0

Closed issues:

• Release 1.0.0 (#78)

Merged pull requests:

• Documenter hotfix (#114) (@PyDataBlog)
• Minor update (#115) (@PyDataBlog)

ParallelKMeans v1.0.0

Diff since v0.2.2

Closed issues:

• Benchmark figure (#79)

Merged pull requests:

• Updated benchmark files. Table & Plot left (#108) (@PyDataBlog)
• Revert "Updated benchmark files. Table & Plot left" (#110) (@PyDataBlog)
• Revert "Revert "Updated benchmark files. Table & Plot left"" (#111) (@PyDataBlog)
• Public release v1.0.0 (#112) (@PyDataBlog)

ParallelKMeans v0.2.2

Diff since v0.2.1

Closed issues:

• Release 0.2.1 (#105)

Merged pull requests:

• CompatHelper: bump compat for "MLJModelInterface" to "1.0" (#106) (@github-actions[bot])
• bump version (#107) (@Arkoniak)

ParallelKMeans v0.2.1

Diff since v0.2.0

Closed issues:

• TagBot trigger issue (#99)

Merged pull requests:

• MiniBatch algorithm draft (#101) (@PyDataBlog)
• Optimise mini (#102) (@PyDataBlog)
• fixed CI (#103) (@Arkoniak)
• Fixed lower boundary (#104) (@Arkoniak)

ParallelKMeans v0.2.0

Diff since v0.1.9

Closed issues:

• Release 0.1.9 (#95)

Merged pull requests:

• CompatHelper: bump compat for "MLJModelInterface" to "0.4" (#97) (@github-actions[bot])

ParallelKMeans v0.1.9

Diff since v0.1.8

Closed issues:

• Extend [compat] MLJModelInterface = "^0.2,^0.3" (#89)
• Not working on this example. Only gives one value (#93)

Merged pull requests:

• CompatHelper: bump compat for "Distances" to "0.10" (#94) (@github-actions[bot])
• Update travis setup (#96) (@PyDataBlog)

ParallelKMeans v0.1.8

Diff since v0.1.7

Closed issues:

• Release 0.1.7 (#77)

Merged pull requests:

• CompatHelper: bump compat for "Distances" to "0.9" (#85) (@github-actions[bot])
• CompatHelper: bump compat for "MLJModelInterface" to "0.3" (#88) (@github-actions[bot])
• version bump (#90) (@Arkoniak)

ParallelKMeans v0.1.7

Diff since v0.1.6

Closed issues:

• Expose the RNG as a hyper-parameter (#66)
• Add binder support for live interactive testing (#73)
• Update MLJ Interface with Yinyang and Coresets algorithm (#74)
• possible test failure in upcoming Julia version 1.5 (#80)

Merged pull requests:

• Support for distance metrics (#75) (@PyDataBlog)
• Improved support of MLJ (#81) (@Arkoniak)

ParallelKMeans v0.1.6

Diff since v0.1.5

Closed issues:

• More generic type support (#22)
• Release 0.1.5 (#68)

Merged pull requests:

• Added Yinyang to experimental (#69) (@PyDataBlog)
• types broadening (#70) (@Arkoniak)
• Coresets (#72) (@Arkoniak)

ParallelKMeans v0.1.5

Diff since v0.1.4

Closed issues:

• Add Yinyang K-Means? (#43)
• Release 0.1.4 (#62)

Merged pull requests:

• Simplify the instructions to get the experimental branch (#63) (@asinghvi17)
• typos and grammar (#64) (@xiaodaigh)
• Yinyang algorithm (#65) (@Arkoniak)
• Draft release of v0.1.5 (#67) (@PyDataBlog)

ParallelKMeans v0.1.4

Diff since v0.1.3

Closed issues:

• Release 0.1.3 (#60)

Merged pull requests:

• Updated MLJ Interface with predict function (#61) (@PyDataBlog)

ParallelKMeans v0.1.3

Diff since v0.1.2

Closed issues:

• Fix broken Git Actions BenchmarkCI (#38)
• [Patch Release v0.1.2] Added Full Elkan & Refactored MLJ Interface (#52)
• Support for distance metrics beyond Euclidian. (#56)

Merged pull requests:

• Attempting to fix broken git actions Benchmark (#51) (@PyDataBlog)
• Testing proposed fix for broken benchmark git actions (#53) (@PyDataBlog)
• Ignoring benchmark/Manifest.toml (#54) (@PyDataBlog)
• simd for faster calculations (#57) (@Arkoniak)
• Release of v0.1.3 (#59) (@PyDataBlog)

ParallelKMeans v0.1.2

Diff since v0.1.1

Closed issues:

• Patch Release 0.1.1 - MLJModels Interface (#49)

Merged pull requests:

• Elkan and refactoring (#46) (@Arkoniak)
• Get both prod and dev branches on the same page. (#50) (@PyDataBlog)

ParallelKMeans v0.1.1

Diff since v0.1.0

Closed issues:

• Correctly set up distinct pages for the two main branches (#26)
• Initial version release (#39)
• [Patch Release v0.1.1]: MLJModels Interface (#47)

Merged pull requests:

• fixed broken compilation warning (#41) (@Arkoniak)
• Mlj port (#42) (@PyDataBlog)
• Mljinterface draft (#44) (@PyDataBlog)
• Experimental (#45) (@PyDataBlog)
• Updated project [compat] with new deps (#48) (@PyDataBlog)

ParallelKMeans v0.1.0

Closed issues:

• Smart init needs to be thoroughly vetted and tested. Currently buggy and unstable (#7)

Merged pull requests:

• CompatHelper: add new compat entry for "StatsBase" at version "0.32" (#1) (@github-actions[bot])
• link clean-up (#2) (@PyDataBlog)
• fix documentation (#3) (@Arkoniak)
• refactoring with multiple dispatch (#4) (@Arkoniak)
• attempting to fix broke travis config (#5) (@PyDataBlog)
• Experimental (#6) (@PyDataBlog)
• add init (#9) (@Arkoniak)
• Get the experimental branch up to date with the current stable version (#12) (@PyDataBlog)
• Fixed kmpp init (#13) (@Arkoniak)
• Experimental (but stable) implementation of single thread (#14) (@PyDataBlog)
• pointwise initial (#15) (@Arkoniak)
• Fix benchmark test (#16) (@PyDataBlog)
• fixed kmpp implementation (#19) (@Arkoniak)
• Betterdocs (#21) (@PyDataBlog)
• Implemented light Elkan version of clustering (#27) (@Arkoniak)
• Experimental (#28) (@PyDataBlog)
• refactored code and added doc strings (#29) (@Arkoniak)
• removed type instability (#30) (@Arkoniak)
• CompatHelper: bump compat for "StatsBase" to "0.33" (#31) (@github-actions[bot])
• Initial hamerly implementation (#32) (@Arkoniak)
• Experimental (#33) (@PyDataBlog)
• Cleandocs (#35) (@PyDataBlog)
• added Hamerly docstring (#36) (@Arkoniak)
• Experimental (#37) (@PyDataBlog)