Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on May 18, 2012

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  5cd7e59

  Copy full SHA for 5cd7e59

• Several fixes in distance and kmeans

  flaviotruzzi

  committed
  44f5015

  Copy full SHA for 44f5015

Commits on May 17, 2012

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  815b0a1

  Copy full SHA for 815b0a1

Commits on May 15, 2012

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  f5d952a

  Copy full SHA for f5d952a

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  bb78003

  Copy full SHA for bb78003

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  6f3243e

  Copy full SHA for 6f3243e

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  b87f329

  Copy full SHA for b87f329

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  4b30a76

  Copy full SHA for 4b30a76

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  42fb43f

  Copy full SHA for 42fb43f

• Improved euclidian distance for kmeans

  flaviotruzzi

  committed
  c7e0b57

  Copy full SHA for c7e0b57

Commits on May 11, 2012

• morpho madness

  nlw0

  committed
  d25789c

  Copy full SHA for d25789c

• using getenv...

  flaviotruzzi

  committed
  410a73a

  Copy full SHA for 410a73a

• Dilation not quite working right

  nlw0

  committed
  1a2761c

  Copy full SHA for 1a2761c

• morphological erosion by multiple directional structuring elements, to remove small points from the venation class

  nlw0

  committed
  3d06f80

  Copy full SHA for 3d06f80

Commits on May 10, 2012

• makefile pro schraudolph.

  nlw0

  committed
  6dffe71

  Copy full SHA for 6dffe71

• Finalmente uma rotina de exp rápida que funciona bem, tirada da biblitoeca do "Paul Mineiro"

  nlw0

  committed
  57d7245

  Copy full SHA for 57d7245

Commits on May 9, 2012

• super-ultra-local segmentation. Does not seem to be totally wrong, but it's slooow, and may not be the best approach, we need osme kind of smoothening, a

  Show description for 2947139
  nlw0

  committed
  2947139

  Copy full SHA for 2947139

• starting new python implementation to test local segmentation version, with mu coefficients changing alongi the image

  nlw0

  committed
  a22e2a4

  Copy full SHA for a22e2a4

Commits on May 8, 2012

• maybe

  

  [email protected]

  committed
  4932208

  Copy full SHA for 4932208

Commits on May 7, 2012

• Kullback–Leibler divergence

  

  [email protected]

  committed
  f4609f7

  Copy full SHA for f4609f7

• Changed some stuffs...

  

  [email protected]

  committed
  744a95a

  Copy full SHA for 744a95a

Commits on May 5, 2012

• Speeding up the EStep by moving the old veross procedure to the body of the EStep... 'inline' was cheating us. Also included an approximate fast exponential

  Show description for 6ea8333
  nlw0

  committed
  6ea8333

  Copy full SHA for 6ea8333

Commits on May 4, 2012

• Otimizando _geral_. Tempo médio de cada iteração E+M ~0.3 seg. com 200x151 pixels, sendo que boa parte da culpa parece ser da função exp(), então parece bom.

  Show description for fcfdd6f
  nlw0

  committed
  fcfdd6f

  Copy full SHA for fcfdd6f

• aparentemente agora está funcionando tudo. faltava calcular o z direito e incluir as probs de cada classe (os "pis")

  nlw0

  committed
  2ba40d1

  Copy full SHA for 2ba40d1

• ...

  flaviotruzzi

  committed
  f07259c

  Copy full SHA for f07259c

Commits on May 3, 2012

• Fixed.

  flaviotruzzi

  committed
  70e5b44

  Copy full SHA for 70e5b44

• Initial...

  flaviotruzzi

  committed
  5ec70c5

  Copy full SHA for 5ec70c5