aboutsummaryrefslogtreecommitdiff
path: root/include
AgeCommit message (Expand)Author
2 daysWent back up to 1986 weightstslil
2 daysQuick sketch of non-determinismtslil
2 daysAnother road bugtslil
2 daysDragon is not separate, fixed concession, allowing playing as b/wtslil
2 daysChanged function back to macrotslil
2 daysThere was a bug in the minimax!tslil
2 daysBetter trained weights, an important oversight in minimax (steps <=)tslil
2 daysSo after a small tweak, it actually plays well!tslil
2 daysSeems tolerable, but not great. Still wont win or lose ???tslil
2 daysTrying alpha-beta pruning. There's a bug somewhere ...tslil
2 daysFixed minimax (!), fixed bugs in tak.ctslil
2 daysTrying minimaxtslil
2 daysCrass errortslil
2 daysWeights?tslil
2 daysWider?tslil
2 daysTrying various things to teach titslil
2 daysHave to account for CPStslil
2 daysWell it works, and it's bad. Many bugs fixed in interimtslil
2 daysClosing in!tslil
2 daysAuto-generating weightstslil
2 daysNew weightstslil
2 daysone output (of course), Pade approximant of logistic for activationtslil
2 daysTrain on data that is close to the end of the game onlytslil
2 daysThis finally works!tslil
2 daysClose in principle, but there;s a bug in ct1997 &/ pptdb!tslil
2 daysThis is the general idea, but i'm sure order is incorrect somewheretslil
2 daysStarting on importing network!tslil
2 daysFixed a small bug in PTN parsing, other stufftslil
2 daysCheck for road wins before flat winstslil
2 daysTabs for indentation, spaces for alignmenttslil
2 daysFixed indentation and some bugs, stats programmetslil
2 daysSet dialect to C99tslil
2 daysIncorrect usage of globals in headertslil
2 daysIn-lined a few things, reworked flat count to iterate only oncetslil
2 daystak.h now tracks win state, optimisations, saved a byte on countstslil
2 daysLine-mode mostly completetslil
2 daysFixed road DFS errorstslil
2 daysBoard printingtslil
2 daysCorrected PTN move parsingtslil
2 daysAdded linenoise library, uninitialised variable + renamingtslil
2 daysFixed orderingtslil
2 daysTransition to single file headertslil
2 daysBetter as single file headertslil
2 daysOBOEs in try_move + logic errorstslil
2 daysWorking on the big wrapper for everythingtslil
2 daysOutput PTN as welltslil
2 daysPTN code looks correct, minor tweaks and header guardstslil
2 daysStarted on PTN stufftslil
2 daysChanged representation, implemented most basic things for tak.htslil
2 daysInittslil