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

Age	Commit message (Collapse)	Author
38 hours	It works, but it's too slowmcu	tslil

38 hours	Starting work on MCU port	tslil

38 hours	Went back up to 1986 weights	tslil

38 hours	Quick sketch of non-determinism	tslil

38 hours	Another road bug	tslil

38 hours	Dragon is not separate, fixed concession, allowing playing as b/w	tslil

38 hours	Changed function back to macro	tslil

38 hours	There was a bug in the minimax!	tslil

38 hours	Better trained weights, an important oversight in minimax (steps <=)	tslil
	I still can't tell whether if ((ply & 1) == 0) this = this - 1.0; is correct.
38 hours	So after a small tweak, it actually plays well!	tslil

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

38 hours	Trying alpha-beta pruning. There's a bug somewhere ...	tslil

38 hours	There are still some bugs somewhere in minimax ...	tslil
	In tinue it doesn't seem to choose a move
38 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
38 hours	Trying minimax	tslil

38 hours	Crass error	tslil

38 hours	Weights?	tslil

38 hours	Wider?	tslil

38 hours	Trying various things to teach ti	tslil

38 hours	Have to account for CPS	tslil

38 hours	Well it works, and it's bad. Many bugs fixed in interim	tslil

38 hours	Closing in!	tslil

38 hours	Auto-generating weights	tslil

38 hours	Generate the correct format directly	tslil

38 hours	New weights	tslil

38 hours	one output (of course), Pade approximant of logistic for activation	tslil

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

38 hours	Working version of this script	tslil

38 hours	This finally works!	tslil

38 hours	Hopefully removed all layer bugs	tslil

38 hours	Trying again with correct data generation	tslil

38 hours	Also this was wrong	tslil

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

38 hours	This is the general idea, but i'm sure order is incorrect somewhere	tslil

38 hours	Starting on importing network!	tslil

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

38 hours	Fixed comments during extraction	tslil

38 hours	Shuffling is important	tslil

38 hours	Magic	tslil

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

38 hours	Closing in on something reasonable for roads	tslil

38 hours	Check for road wins before flat wins	tslil

38 hours	Searching for a good representation	tslil

38 hours	Small tweaks	tslil

38 hours	On the hunt for a better representation	tslil

38 hours	Experimenting with training nn	tslil

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

38 hours	Gotta go fast	tslil

38 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
38 hours	Tabs for indentation, spaces for alignment	tslil