Commits

Breadcrumbs

History for

onmaster

User selector

All users

All time

Commit history

Commits on Oct 13, 2018

• Weight parent also when applying child valueweight

  lightvector

  committed
  bbf4d34

  Copy full SHA for bbf4d34

Commits on Oct 8, 2018

• Move external libs to external, clarify license for sha, fix some includes

  lightvector

  committed
  08949e0

  Copy full SHA for 08949e0

Commits on Oct 7, 2018

• Implement NCHW input features

  lightvector

  committed
  7cda2b2

  Copy full SHA for 7cda2b2

Commits on Oct 6, 2018

• Merge variable board size prep

  lightvector

  committed
  f02018c

  Copy full SHA for f02018c

Commits on Oct 5, 2018

• Fix minor cfg bug, and a display bug when showing match moves

  lightvector

  committed
  83852bb

  Copy full SHA for 83852bb

• Pull useful parts of match into play lib

  lightvector

  committed
  9b125e9

  Copy full SHA for 9b125e9

Commits on Oct 4, 2018

• Make rand constructor thread-safe, misc

  lightvector

  committed
  34f9124

  Copy full SHA for 34f9124

• Expose debugskipneuralnet in cfg

  lightvector

  committed
  22f0dab

  Copy full SHA for 22f0dab

Commits on Sep 25, 2018

• Begin support for variable board sizes in search and neural net

  lightvector

  committed
  057dce1

  Copy full SHA for 057dce1

Commits on Sep 23, 2018

• Embed git hash into main program

  lightvector

  committed
  301c746

  Copy full SHA for 301c746

• Clean up some todos

  lightvector

  committed
  5da5dfd

  Copy full SHA for 5da5dfd

Commits on Sep 19, 2018

• Log different nn loads

  lightvector

  committed
  f68a4df

  Copy full SHA for f68a4df

• Implement nhwc option

  lightvector

  committed
  b7d421f

  Copy full SHA for b7d421f

Commits on Sep 17, 2018

• Error checking, bugfixes, attempt to use tensor cores

  lightvector

  committed
  35cd54b

  Copy full SHA for 35cd54b

• Implement fp16

  lightvector

  committed
  775b95d

  Copy full SHA for 775b95d

Commits on Sep 9, 2018

• Clean up and rename move model stuff

  lightvector

  committed
  33567a6

  Copy full SHA for 33567a6

Commits on Aug 26, 2018

• Move model prob exponent and visits exponent

  lightvector

  committed
  37092a2

  Copy full SHA for 37092a2

Commits on Aug 19, 2018

• Finish up model integration for value weighting

  lightvector

  committed
  f20c2d1

  Copy full SHA for f20c2d1

Commits on Aug 15, 2018

• Fix bug with truncated gaussian in setup

  lightvector

  committed
  84549f8

  Copy full SHA for 84549f8

Commits on Aug 9, 2018

• Try option to use parent average for fpu instead of parent nn value

  lightvector

  committed
  e8185be

  Copy full SHA for e8185be

Commits on Jul 29, 2018

• Various changes to prepare for cuda backend

  lightvector

  committed
  00bb99c

  Copy full SHA for 00bb99c

Commits on Jul 27, 2018

• Move tensorflow behind a generalizable interface

  lightvector

  committed
  ef5b519

  Copy full SHA for ef5b519

Commits on Jul 25, 2018

• Logging, sgf writing bugfix, temperature control

  lightvector

  committed
  ea4ca30

  Copy full SHA for ea4ca30

Commits on Jul 23, 2018

• Simplify match config spec

  lightvector

  committed
  bcfaa7d

  Copy full SHA for bcfaa7d

• Fix up tensorflow session stuff

  lightvector

  committed
  4c54fda

  Copy full SHA for 4c54fda

Commits on Jul 22, 2018

• Fix a bunch of compile stuff

  lightvector

  committed
  c3fff04

  Copy full SHA for c3fff04

Commits on Jul 21, 2018

• Komi rand, handicap games, move rand seed out of search params

  lightvector

  committed
  c58cb03

  Copy full SHA for c58cb03

• Account for jigo

  lightvector

  committed
  5366945

  Copy full SHA for 5366945

Commits on Jul 20, 2018

• Factor out more stuff to setup

  lightvector

  committed
  878379c

  Copy full SHA for 878379c

• Config parser stuff, factor stuff out, other misc stuff

  lightvector

  committed
  c2ba215

  Copy full SHA for c2ba215

Loading