ctak - An implementation of Tak and a computer opponent in C

Age	Commit message (Collapse)	Author
43 hours	Change the training data generation a little	tslil clingman
	Although it pains me to say it, ``label smoothing'' appears to be actually work. I'm also currently experimenting with training simply against _all_ games, instead of only bot matches. Once the training finishes i'll pit cttei against itself with old and new weights, hopefully there'll be a noticeable improvement.
43 hours	TEI interface working!	tslil clingman

43 hours	Tried some naive iterative deepening. Work on TEI interface next	tslil clingman
	If TEI is implemented, then i could make use of Morten's racetrack (https://github.com/MortenLohne/racetrack) and develop a quantitative measure of the bot's performance. This is the current priority.
43 hours	Just some #weightgoals ;)	tslil clingman
	It turns out that while i was training on a 0/1 classification problem, i was using 2*eval - 1. Training using this function instead, and on bot-dominated game choices (chosen_player in extract.sh) seems to have given a better evaluation function. At the least, Morten's swindle doesn't work anymore.
43 hours	Syntax errors, small tweak to training data generation	tslil clingman

43 hours	Added license information!	tslil clingman

43 hours	Added tunable search depth and self-play	tslil clingman

43 hours	Toying with symmetrising data	tslil clingman

43 hours	Went back up to 1986 weights	tslil

43 hours	Changed function back to macro	tslil

43 hours	Seems tolerable, but not great. Still wont win or lose ???	tslil

43 hours	Fixed minimax (!), fixed bugs in tak.c	tslil
	With minimax of depth 1 the evaluation function seems alright with the current method of training and generating weights
43 hours	Trying minimax	tslil

43 hours	Weights?	tslil

43 hours	Wider?	tslil

43 hours	Trying various things to teach ti	tslil

43 hours	Generate the correct format directly	tslil

43 hours	Train on data that is close to the end of the game only	tslil

43 hours	Hopefully removed all layer bugs	tslil

43 hours	Trying again with correct data generation	tslil

43 hours	Close in principle, but there;s a bug in ct1997 &/ pptdb!	tslil

43 hours	Two inputs to model, stacks and flat counts	tslil

43 hours	Shuffling is important	tslil

43 hours	Fixed a small bug in PTN parsing, other stuff	tslil

43 hours	Closing in on something reasonable for roads	tslil

43 hours	Searching for a good representation	tslil

43 hours	Small tweaks	tslil

43 hours	On the hunt for a better representation	tslil

43 hours	Experimenting with training nn	tslil

43 hours	Separate concerns in pptdb, remove legacy in extract.sh	tslil

43 hours	Gotta go fast	tslil

43 hours	First attempt at training data + exclude draws + list overflows too	tslil
	Take the output and do grep -Fvxf output_file input_file to drop the bad games from consideration
43 hours	Tabs for indentation, spaces for alignment	tslil

43 hours	Fixed indentation and some bugs, stats programme	tslil