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

Age	Commit message (Collapse)	Author
3 days	Went back up to 1986 weights	tslil

3 days	Dragon is not separate, fixed concession, allowing playing as b/w	tslil

3 days	Changed function back to macro	tslil

3 days	There was a bug in the minimax!	tslil

3 days	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.
3 days	So after a small tweak, it actually plays well!	tslil

3 days	Seems tolerable, but not great. Still wont win or lose ???	tslil

3 days	Trying alpha-beta pruning. There's a bug somewhere ...	tslil

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

3 days	Weights?	tslil

3 days	Wider?	tslil

3 days	Trying various things to teach ti	tslil

3 days	Have to account for CPS	tslil

3 days	Well it works, and it's bad. Many bugs fixed in interim	tslil

3 days	Closing in!	tslil

3 days	Auto-generating weights	tslil

3 days	Generate the correct format directly	tslil

3 days	one output (of course), Pade approximant of logistic for activation	tslil

3 days	Train on data that is close to the end of the game only	tslil

3 days	Working version of this script	tslil

3 days	Hopefully removed all layer bugs	tslil

3 days	Trying again with correct data generation	tslil

3 days	Also this was wrong	tslil

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

3 days	Two inputs to model, stacks and flat counts	tslil

3 days	Shuffling is important	tslil

3 days	Magic	tslil

3 days	Fixed a small bug in PTN parsing, other stuff	tslil

3 days	Closing in on something reasonable for roads	tslil

3 days	Searching for a good representation	tslil

3 days	Small tweaks	tslil

3 days	On the hunt for a better representation	tslil

3 days	Experimenting with training nn	tslil

3 days	Separate concerns in pptdb, remove legacy in extract.sh	tslil

3 days	Gotta go fast	tslil

3 days	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
3 days	Tabs for indentation, spaces for alignment	tslil

3 days	Fixed indentation and some bugs, stats programme	tslil

3 days	Set dialect to C99	tslil

3 days	Added help, trimmed print_info	tslil

3 days	tak.h now tracks win state, optimisations, saved a byte on counts	tslil

3 days	Silly mistake	tslil

3 days	Line-mode mostly complete	tslil

3 days	Working on line-mode front-end	tslil

3 days	Fixed road DFS errors	tslil

3 days	Board printing	tslil

3 days	Added linenoise library, uninitialised variable + renaming	tslil

3 days	Transition to single file header	tslil