Search found 252 matches

by brianr
Tue Sep 18, 2018 5:08 pm
Forum: Computer Chess Club: General Topics
Topic: If you could buy any single CPU system for chess . . .
Replies: 32
Views: 1330

Re: If you could buy any single CPU system for chess . . .

Milos is probably talking about the cost of building a system himself. I don't have those skills or knowledge. If your footer actually means you have a PhD, then you are smart enough to watch a youtube video and build your own system. Theres very few things to screw up when building. It is less a q...
by brianr
Sun Sep 16, 2018 12:57 am
Forum: Computer Chess Club: General Topics
Topic: Chess.com 2018 computer chess championship
Replies: 260
Views: 18036

Re: Chess.com 2018 computer chess championship

AndrewGrant wrote:
Sun Sep 16, 2018 12:55 am
That jump assumes that all engines have NUMA support. It is my understanding that only Stockfish, Houdini, Texel, and Ethereal have such support. I'm tempted to add Komodo into that too, based on some chats with Mark, but I don't know that first hand.
Crafty has NUMA support
by brianr
Mon Sep 03, 2018 5:19 pm
Forum: Computer Chess Club: General Topics
Topic: Cutechess-cli.exe with -noswap option
Replies: 4
Views: 211

Re: Cutechess-cli.exe with -noswap option

That works!
Thanks.

I thought I had tried that (the latest version), but sometimes you start flailing around and don't know what you are doing.
Been one of those days for me.

Thanks again,
Brian
by brianr
Mon Sep 03, 2018 4:29 pm
Forum: Computer Chess Club: General Topics
Topic: Cutechess-cli.exe with -noswap option
Replies: 4
Views: 211

Re: Cutechess-cli.exe with -noswap option

Thanks.

Naturally, my Qt5Core.dll does not have a required entry point...

Might you have the Qt .dll files needed also?

Thanks,
Brian
by brianr
Mon Sep 03, 2018 1:43 pm
Forum: Computer Chess Club: General Topics
Topic: Cutechess-cli.exe with -noswap option
Replies: 4
Views: 211

Cutechess-cli.exe with -noswap option

Anybody have a .exe or a link to the version of cli that supports the -noswap option?

I've looked for quite some time and cannot find one. There are several older versions, but not with -noswap

Thanks,
Brian

PS -noswap not in my Linux version of 1.0 either...weird
What am I missing?
by brianr
Sat Sep 01, 2018 12:06 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Learning time growing exponentially with number of training examples
Replies: 34
Views: 1337

Re: Learning time growing exponentially with number of training examples

Hi hi hi. It might even be true that a network with bad learning properties or with bugs performs best for it generalizes better. Working on Tinker I found found several times that a version with clear bugs actually played better. But, as conflicted as I was, I finally came around to working only f...
by brianr
Sat Sep 01, 2018 12:00 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Learning time growing exponentially with number of training examples
Replies: 34
Views: 1337

Re: Learning time growing exponentially with number of training examples

There are many network architectures. Why should a residual network be the best. Perhaps great many variants to try. But unfortunately testing them takes so much time. So my strategy is to start with smallest ones first. For it should take less time to test or to find out it is not working and also...
by brianr
Sat Sep 01, 2018 10:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Learning time growing exponentially with number of training examples
Replies: 34
Views: 1337

Re: Learning time growing exponentially with number of training examples

Slow progress. 70000 examples and each epoch taking ten minutes. When progress stops I change network and it starts all over again. Maybe adding more training examples might help if progress stops. I don't know. Haven't tested that. Or perhaps I better use 10000 examples or less to evaluate a netwo...
by brianr
Fri Aug 31, 2018 1:11 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Learning time growing exponentially with number of training examples
Replies: 34
Views: 1337

Re: Learning time growing exponentially with number of training examples

It's also questionable if a network can truly unlearn wrong information. It seems logical on its face, yet Deepmind found that networks that were started from analysis of expert games topped out lower than a network starting from scratch. If they could really unlearn things completely, then both sh...
by brianr
Fri Aug 31, 2018 12:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Learning time growing exponentially with number of training examples
Replies: 34
Views: 1337

Re: Learning time growing exponentially with number of training examples

I am currently training a very small endgame (KQvK) net that is "32x4" in terms of the AlphaZero approach with around 1.15 million parameters. The smallest run that I do is with 144,000 games, where each starting position and every move in each game is a sample. There are about 1.7 million samples. ...