Code

Adjusted Mutual Information#402

Code

Merged

robertlayton merged 38 commits intoscikit-learn:masterscikit-learn/scikit-learn:masterfrom

robertlayton:amirobertlayton/scikit-learn:amiCopy head branch name to clipboard

Nov 10, 2011

Commits

Commits on Oct 14, 2011

• Trying to fix NaN errors, but its not working. Pushing to work on it later.

  robertlayton

  committed
  2eb9833

  Copy full SHA for 2eb9833

Commits on Oct 15, 2011

• Mutual information now works (tested!)

  robertlayton

  committed
  c3d3906

  Copy full SHA for c3d3906

Commits on Oct 19, 2011

• AMI now works, and has been tested against the matlab code (test based on this to come!)

  robertlayton

  committed
  9f05c11

  Copy full SHA for 9f05c11

• Remove phantom double v-measure !?

  robertlayton

  committed
  e473f70

  Copy full SHA for e473f70

• Added tests. There are two errors, but I'm going to bed. I'll fix them in the morning.

  robertlayton

  committed
  2d8677f

  Copy full SHA for 2d8677f

Commits on Oct 20, 2011

• Merge branch 'master' into ami

  Show description for af91582
  robertlayton

  committed
  af91582

  Copy full SHA for af91582

• Merge branch 'ami' of github.com:robertlayton/scikit-learn into ami

  robertlayton

  committed
  770020f

  Copy full SHA for 770020f

• - AMI in the cluster examples

  Show description for 193ac02
  robertlayton

  committed
  193ac02

  Copy full SHA for 193ac02

• Higher level import for ami_score

  robertlayton

  committed
  867ec2f

  Copy full SHA for 867ec2f

• There is an overflow problem. It can be reproduced with the plot_adjusted_for_chance_measures.py example

  Show description for 18b9a78
  robertlayton

  committed
  18b9a78

  Copy full SHA for 18b9a78

Commits on Oct 21, 2011

• Narrative doc, and I think I fixed the overflow issue (more tests to come)

  robertlayton

  committed
  aaf5c23

  Copy full SHA for aaf5c23

• Fixed logs to match the matlab code results.

  Show description for fb0c3f8
  robertlayton

  committed
  fb0c3f8

  Copy full SHA for fb0c3f8

• Test now tests a much larger array

  robertlayton

  committed
  db845fa

  Copy full SHA for db845fa

• Test actually does what I meant it to do, and works sufficiently

  robertlayton

  committed
  77ec530

  Copy full SHA for 77ec530

• Fixed this example. Tested the others (they worked!)

  robertlayton

  committed
  52fc4c5

  Copy full SHA for 52fc4c5

• pep8 and pyflakes

  robertlayton

  committed
  c021b59

  Copy full SHA for c021b59

Commits on Oct 22, 2011

• Merge remote-tracking branch 'robertlayton/ami' into robertlayton-ami

  ogrisel

  committed
  3052291

  Copy full SHA for 3052291

Commits on Oct 23, 2011

• measure runtimes for various clustering metrics in adjusted for chance example

  ogrisel

  committed
  b7d7642

  Copy full SHA for b7d7642

• FIX warnings by avoiding 0.0 values in the log + cosmit

  ogrisel

  committed
  3845fd8

  Copy full SHA for 3845fd8

• Merge pull request #3 from ogrisel/robertlayton-ami

  Show description for b89b6d0
  robertlayton

  committed
  b89b6d0

  Copy full SHA for b89b6d0

• Optimising the expected mutual information code

  Show description for 77ff098
  robertlayton

  committed
  77ff098

  Copy full SHA for 77ff098

Commits on Oct 29, 2011

• Adding old version of EMI, as I'm about to change it

  robertlayton

  committed
  13a48df

  Copy full SHA for 13a48df

Commits on Nov 2, 2011

• This version doesn't work either. I am uploading for historical sake.

  Show description for ccc3cad
  robertlayton

  committed
  ccc3cad

  Copy full SHA for ccc3cad

• Initial usage of gammaln. Not yet tested

  robertlayton

  committed
  fb6fe36

  Copy full SHA for fb6fe36

• Merge branch 'master' of https://github.com/scikit-learn/scikit-learn into ami

  robertlayton

  committed
  a786bdc

  Copy full SHA for a786bdc

Commits on Nov 5, 2011

• Still overflows, but the closest so far. Using gammaln

  robertlayton

  committed
  75bf5b0

  Copy full SHA for 75bf5b0

• It works! Still have some optimisation to do, but it works for larger arrays

  robertlayton

  committed
  01ae29e

  Copy full SHA for 01ae29e

Commits on Nov 6, 2011

• Moved start and finish outside of loop

  robertlayton

  committed
  f407372

  Copy full SHA for f407372

Commits on Nov 7, 2011

• comments, pep8 and pyflakes

  robertlayton

  committed
  88adf3b

  Copy full SHA for 88adf3b

• Merge branch 'master' of https://github.com/scikit-learn/scikit-learn into ami

  robertlayton

  committed
  9f6ca7f

  Copy full SHA for 9f6ca7f

• ami_score -> adjusted_mutual_info_score

  Show description for 2fa1052
  robertlayton

  committed
  2fa1052

  Copy full SHA for 2fa1052

• ami_score -> adjusted_mutual_info_score

  Show description for e51e4d6
  robertlayton

  committed
  e51e4d6

  Copy full SHA for e51e4d6

Commits on Nov 8, 2011

• "What's new?" AMI!

  robertlayton

  committed
  66f7a0b

  Copy full SHA for 66f7a0b

Commits on Nov 10, 2011

• Merge branch 'ami' of https://github.com/robertlayton/scikit-learn into ami

  Show description for b9685ff
  robertlayton

  committed
  b9685ff

  Copy full SHA for b9685ff

• mutual_information_score -> mutual_info_score

  robertlayton

  committed
  f26ed76

  Copy full SHA for f26ed76

• and in plot_adjusted example (mutual_info_score)

  robertlayton

  committed
  17ee6c4

  Copy full SHA for 17ee6c4

• Merge branch 'master' of https://github.com/scikit-learn/scikit-learn into ami

  robertlayton

  committed
  ff0dbf9

  Copy full SHA for ff0dbf9

• cosmit

  robertlayton

  committed
  118e8bd

  Copy full SHA for 118e8bd