LCZero update

Discussion of anything and everything relating to chess playing software and machines.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Locked
CheckersGuy
Posts: 273
Joined: Wed Aug 24, 2016 7:49 pm

Re: LCZero update

Post by CheckersGuy » Sat Mar 17, 2018 12:53 am

Probably need to wait at least a week before those games become watchable :P Watching some of the training games hurt my eyes :lol:

User avatar
CMCanavessi
Posts: 835
Joined: Thu Dec 28, 2017 3:06 pm
Location: Argentina

Re: LCZero update

Post by CMCanavessi » Sat Mar 17, 2018 1:06 am

pferd wrote:
CMCanavessi wrote:
pferd wrote: Does anybody know how strength increases when the number of playouts is raised?
Playouts play a HUGE role in strenght. All the games I've been running use the default 800, which makes it rather weak but is good for the training purposes, as it provides more games/time. I see you use 50000, that's a nice number and should make it much stronger.
I upped the playouts to 150000 (equals 30-40s/move on my machine) and had a quick match against TSCP 1.81b which was using 1 min/game. The network in use was 1aa02cb1. The game with Leela using the white pieces did not look too bad for :

[pgn]
[Event "Computer Chess Game"]
[Site "brutus"]
[Date "2018.03.16"]
[Round "1"]
[White "lczero"]
[Black "tscp"]
[Result "0-1"]
[TimeControl "60"]
[Annotator "1. -0.04 1... -0.30"]

1.d4
{-0.04/25}
1...d5
{-0.30/6 2.0}
2.Nf3
{-0.10/26 34}
2...Nf6
{-0.25/6 1.9}
3.c3
{-0.12/26 33}
3...Nc6
{+0.02/6 1.9}
4.b4
{-0.34/26 39}
4...e6
{+0.27/6 1.8}
5.a4
{-0.18/26 38}
5...Bd6
{+0.89/5 1.7}
6.Qd2
{-0.01/26 40}
6...O-O
{+1.02/6 1.7}
7.Qb2
{-0.05/26 41}
7...Bd7
{+0.92/6 1.6}
8.b5
{+0.03/26 41}
8...Ne7
{+0.81/6 1.6}
9.Ne5
{-0.04/26 41}
9...Bxe5
{+1.47/6 1.5}
10.dxe5
{-0.14/26 40}
10...Ng4
{+1.48/6 1.5}
11.Na3
{-0.49/26 39}
11...Nxe5
{+1.92/5 1.4}
12.g3
{-0.53/26 34}
12...c5
{+2.06/5 1.4}
13.f4
{+0.05/26 37}
13...Ng4
{+1.94/6 1.3}
14.e3
{+0.23/26 40}
14...e5
{+2.00/5 1.3}
15.Be2
{+0.17/26 40}
15...Qa5
{+1.83/6 1.2}
16.Nb1
{+0.84/26 40}
16...exf4
{+2.06/5 1.2}
17.exf4
{+1.21/26 42}
17...Rfe8
{+1.93/5 1.2}
18.Nd2
{+1.73/26 40}
18...Nf5
{+1.82/5 1.1}
19.Nb3
{+2.56/26 41}
19...Qb6
{+1.87/5 1.1}
20.O-O
{+1.40/26 40}
20...c4+
{+3.67/5 1.1}
21.Nd4
{+1.31/26 42}
21...Nxd4
{+3.67/5 1.0}
22.Bxg4
{-1.92/26 41}
22...Ne2+
{+4.80/5 1.0}
23.Kg2
{-1.64/26 40}
23...Bxg4
{+4.92/5 0.9}
24.Re1
{-1.53/26 40}
24...d4
{+5.55/5 0.9}
25.cxd4
{-2.47/26 40}
25...c3
{+5.59/4 0.9}
26.Qb3
{-2.54/26 39}
26...Qxd4
{+5.96/4 0.9}
27.Rd1
{-3.52/26 40}
27...Qe4+
{+99.93/5 0.4}
28.Kf2
{-28.77/20 2.2}
28...Qf3+
{+99.95/4 0.1}
29.Ke1
{-36.14/19 0.7}
29...Nxc1+
{+99.97/3 0.1}
30.Qe6
{-41.03/18 0.5}
30...Rxe6#
{Xboard adjudication: Checkmate}
0-1
[/pgn]

The reverse game however, was very hard to watch with Leela blundering a knight on move 2.

[pgn][Event "Computer Chess Game"]
[Site "brutus"]
[Date "2018.03.16"]
[Round "2"]
[White "tscp"]
[Black "lczero"]
[Result "1-0"]
[TimeControl "60"]
[Annotator "1. +0.13 1... +0.26"]

1.e4
{+0.13/6}
1...Nc6
{+0.26/25 36}
2.d4
{+0.22/6 1.9}
2...Nxd4
{+0.28/26 37}
3.Qxd4
{+2.30/6 1.9}
3...Nf6
{-0.09/26 31}
4.e5
{+3.00/5 1.8}
4...Ng8
{-0.74/26 38}
5.Nc3
{+2.90/5 1.7}
5...f6
{-0.47/26 36}
6.Bc4
{+2.93/5 1.7}
6...b5
{-0.27/26 38}
7.Bxg8
{+5.41/5 1.6}
7...Bb7
{-1.66/26 38}
8.Bd5
{+5.55/5 1.6}
8...c6
{-1.40/26 38}
9.Bf3
{+5.52/5 1.5}
9...Qb8
{-1.74/26 35}
10.Bf4
{+5.93/5 1.5}
10...d6
{-1.74/26 41}
11.exd6
{+6.05/5 1.4}
11...exd6
{-1.75/26 40}
12.O-O-O
{+6.24/4 1.4}
12...h5
{-2.49/26 39}
13.Qe4+
{+6.73/4 1.3}
13...Kf7
{-1.77/26 42}
14.Nxb5
{+6.82/5 1.3}
14...d5
{-3.61/26 40}
15.Qf5
{+7.32/5 1.2}
15...Qd8
{-3.88/26 40}
16.Nc7
{+7.88/5 1.2}
16...Bc5
{-3.90/26 40}
17.Bxh5+
{+9.43/4 1.2}
17...Kg8
{-7.72/26 40}
18.Qe6+
{+9.59/5 1.1}
18...Kh7
{-7.84/26 41}
19.Bf7
{+9.59/4 1.1}
19...g5
{-6.83/26 38}
20.Qf5+
{+13.27/5 1.0}
20...Kg7
{-8.59/26 39}
21.Qg6+
{+13.32/5 1.0}
21...Kf8
{-12.91/26 34}
22.Ne6+
{+13.37/5 1.0}
22...Ke7
{-13.06/26 33}
23.Nxd8
{+16.17/5 0.9}
23...Bxf2
{-13.17/26 32}
24.Qg7
{+17.10/5 0.9}
24...Be3+
{-13.91/25 36}
25.Bxe3
{+21.65/5 0.9}
25...Raxd8
{-14.22/25 37}
26.Bxd5+
{+23.96/4 0.9}
26...Kd6
{-17.83/25 36}
27.Bg8+
{+23.96/4 0.8}
27...Ke5
{-18.43/25 34}
28.Bd4+
{+25.32/5 0.8}
28...Rxd4
{-17.34/25 33}
29.Re1+
{+99.93/5 0.4}
29...Kf4
{-23.48/24 16}
30.Qxf6+
{+99.95/4 0.1}
30...Kg4
{-26.04/23 1.4}
31.Qf3+
{+99.97/3 0.1}
31...Kh4
{-40.26/18 1.1}
32.Qh3#
{Xboard adjudication: Checkmate}
1-0
[/pgn]

Maybe I was bit too optimistic, but let's see where this will go. Interesting days ahead for us.
I also matched it against TSCP, but it's still too much for Leela. It's 1700 elo and it will take a couple of weeks to reach that. i did pair it against the latest Sabrina (around 1350 elo) and it was drawing almost every match. You may try that one.
Follow my tournament and some Leela gauntlets live at http://twitch.tv/ccls

User avatar
CMCanavessi
Posts: 835
Joined: Thu Dec 28, 2017 3:06 pm
Location: Argentina

Re: LCZero update

Post by CMCanavessi » Sat Mar 17, 2018 1:07 am

CheckersGuy wrote:Probably need to wait at least a week before those games become watchable :P Watching some of the training games hurt my eyes :lol:
Yep, but training games are not meant to be good yet, they are just meant to be fast. That's why they are played with a crippled Leela.
Follow my tournament and some Leela gauntlets live at http://twitch.tv/ccls

User avatar
CMCanavessi
Posts: 835
Joined: Thu Dec 28, 2017 3:06 pm
Location: Argentina

Re: LCZero update

Post by CMCanavessi » Sat Mar 17, 2018 1:06 pm

Hi guys, a new network was generated earlier today (Generation 9) so here's a new package with all the files needed to use the engine to help with the training or play against it/in tournaments with a GUI.

You can grab it from here: http://www.mediafire.com/file/21k29oddu20lpz3/LCZero.7z

After downloading the package, unpack it and run the file "Play.bat". It will run a self-tuning routine so it runs as fast as possible in your hardware. That will take a couple of minutes and then it won't be needed anymore, it will open instantly like every other engine.

If you want to help train the new network, run the file "Setup_Training.bat" and follow the instructions given, you'll need to enter a username and password (whatever you want) and your "account" will be automatically created and automatically used after that. Just run Train.bat next time you want to help the training process.


This version I've packaged is a dev version with time control support, so it will always play at full strenght with the given time it has left. Remember that this engine uses your GPU (Video Card) and not your CPU, so having a fast cpu with a lot of cores will not help at all but having a fast modern video card will certainly make the engine stronger.

Happy testing!
Follow my tournament and some Leela gauntlets live at http://twitch.tv/ccls

CheckersGuy
Posts: 273
Joined: Wed Aug 24, 2016 7:49 pm

Re: LCZero update

Post by CheckersGuy » Sat Mar 17, 2018 1:55 pm

CMCanavessi wrote:
CheckersGuy wrote:Probably need to wait at least a week before those games become watchable :P Watching some of the training games hurt my eyes :lol:
Yep, but training games are not meant to be good yet, they are just meant to be fast. That's why they are played with a crippled Leela.
True. However, over at LeelaZero (Go-Engine) training games became watchable after some time despite the randomness that is used to encourage exploration. So I assume that things like blundering a knight on move 2 wont happen :lol:

Nay Lin Tun
Posts: 514
Joined: Mon Jan 16, 2012 5:34 am

Re: LCZero update

Post by Nay Lin Tun » Sun Mar 18, 2018 2:19 am

A kind of random play, quite similar to chimpanzee in chessmaster 8000 (i think) with elo rating '1'.

Jesse Gersenson
Posts: 575
Joined: Sat Aug 20, 2011 7:43 am
Contact:

Re: LCZero update

Post by Jesse Gersenson » Sun Mar 18, 2018 2:33 am

gladius wrote:We have kicked off the reinforcement learning phase, where the network starts from random, and learns from self-play games.

The server is up at http://162.217.248.187/
Hi Gary, when I visit that page it is not clear how I can contribute GPU time to the project. the page is missing a 'call to action' button. add a green button which says 'Get involved' which links to the client software and includes instructions.

In the current .7z file LCZero/Setup_Training.bat includes a typo:
echo Tun.bat will be started once you press Enter
should be
echo Train.bat will be started once you press Enter

User avatar
CMCanavessi
Posts: 835
Joined: Thu Dec 28, 2017 3:06 pm
Location: Argentina

Re: LCZero update

Post by CMCanavessi » Sun Mar 18, 2018 2:49 am

Jesse Gersenson wrote:
gladius wrote:We have kicked off the reinforcement learning phase, where the network starts from random, and learns from self-play games.

The server is up at http://162.217.248.187/
Hi Gary, when I visit that page it is not clear how I can contribute GPU time to the project. the page is missing a 'call to action' button. add a green button which says 'Get involved' which links to the client software and includes instructions.

In the current .7z file LCZero/Setup_Training.bat includes a typo:
echo Tun.bat will be started once you press Enter
should be
echo Train.bat will be started once you press Enter
Yes, the website is kind of a mess because things are changing all the time. If you check a couple of messages above, I uploaded a package ready to use.
Follow my tournament and some Leela gauntlets live at http://twitch.tv/ccls

Werewolf
Posts: 1191
Joined: Thu Sep 18, 2008 8:24 pm

Re: LCZero update

Post by Werewolf » Sun Mar 18, 2018 8:39 am

Is a GeForce 10, 1060 any good for training?

I have one of those. I also have an older Quadro 5000 but it's not as powerful.

noobpwnftw
Posts: 340
Joined: Sun Nov 08, 2015 10:10 pm

Re: LCZero update

Post by noobpwnftw » Sun Mar 18, 2018 10:22 am

Werewolf wrote:Is a GeForce 10, 1060 any good for training?

I have one of those. I also have an older Quadro 5000 but it's not as powerful.
It doesn't have to be powerful, currently it just runs on the previously created network and produce game results, upload them and then they run the training based on everyone's results and create a new network.

Locked