datamllab / rlcard / 360

Builds Branch Commit Type Ran Committer Via Coverage
360 master Merge pull request #40 from ruzhwei/master Holdem multiplayer utils push web-flow travis-ci pending completion   set done
359 master Merge pull request #41 from billmuch/performance-improvement performance improvement for step_back push web-flow travis-ci pending completion   set done
358 master Update utils.py Pull #40 web-flow travis-ci pending completion   set done
357 master Wish this is the last modify Pull #40 web-flow travis-ci pending completion   set done
356 master Update utils.py Pull #40 web-flow travis-ci pending completion   set done
355 master Update test_holdem_utils.py Pull #40 web-flow travis-ci pending completion   set done
354 master Modify multi-player holdem utils Pull #40 web-flow travis-ci pending completion   set done
353 master Modify Multi-player Holdem Utils Pull #40 web-flow travis-ci pending completion   set done
352 performance-improvement performance improvement for step_back Tried cfr on the doudizhu and find 78.8% time was on the deepcopy which is used to record objects for later step_back use. The main changes here are: 1. remove deepcopy 2. for some simple objects like gam... Pull #41 web-flow travis-ci pending completion   set done
351 master Multiplayer holdem utils Pull #40 web-flow travis-ci pending completion   set done
350 master Merge pull request #36 from billmuch/bugfix fix bug: get wrong actions after step() and step_back() in Doudizhu. push web-flow travis-ci pending completion   set done
349 bugfix fix bug: get wrong actions after step() and step_back() in Doudizhu. Doudizhu.game.round.greater_player refered to the same object of doudizhu.game.players[current_player] before step_back but are different objects after step_back() because st... Pull #36 web-flow travis-ci pending completion   set done
348 master Merge pull request #35 from billmuch/bugfix fix bugs and add unittest for them push web-flow travis-ci pending completion   set done
346 master Merge pull request #34 from billmuch/bugfix payoff returned by doudizhu.judger.judge_payoffs should be numpy.ndarray instead of a list push web-flow travis-ci pending completion   set done
345 bugfix Exception when executing cfr_agent.py:75 "state_utility += action_prob * utility" while try cfr_agent with doudizhu. utility in doudizhu (returned in rlcard.doudizhu.judger.py:85) was a list, exception occured when plus it with the state_utility w... Pull #34 web-flow travis-ci pending completion   set done