TalkChess.com

WinPooh wrote: ↑

Wed Jun 05, 2019 11:06 am

Sorry for off-topic, but could anybody give me a link to lc0 network (weights file) which played in TCEC-15 Superfinal? As I know it is not a regular net available from the official lc0 site.

Direct link: http://lc0.org/t40-t8-610
Link to the directory: http://lc0.org/contrib

There is an exception about private software. If you make a modifications for a particular user (or company) and don't release to public, you are free to do that without having to share either binary or sources with the world. I think there never is an obligation to share source with the world. Jus...

GPL allows anyone to make modifications without releasing source code. What GPL doesn't allow is distributing (in GPLv3 terms, "conveying") binary without making source available. Distribution is not only putting it into publicly available place, but also e.g. sending physical disks by mail by reque...

It may be not that bad. Someone is having a very good progress with making an Android version (https://github.com/LeelaChessZero/lc0/pull/848), and I expect that it's not that harder to do for iOS. Probably takes a dozen of hours of free time for a knowledgeable person. Most probably it will be Open...

I don't think it's right to generalize Lc0 weaknesses to weaknesses of all NN engines. Yes, Lc0 has weak spots, but I don't think they are inherent to neural networks, it's just something to fix But for now Lc is all we have. Do you have any reason to believe that A0 does not have the same weakness...

I don't think it's right to generalize Lc0 weaknesses to weaknesses of all NN engines. Yes, Lc0 has weak spots, but I don't think they are inherent to neural networks, it's just something to fix (either in Leela, or it's possible that other engines will be first to address that). For different kinds...

You can enable "Verbose Move Stats" in engine options (or through command line, --verbose-move-stats).

It's at least intended that roundrobin works the best when all GPUs are the same, and for multiplexing it's not that strict.

I wrote a blog post about backends recently: http://blog.lczero.org/2019/04/backend- ... l?m=1#more

In the LC0 folder there's a file CUDNN.txt that contains the "NVIDIA SOFTWARE LICENSE AGREEMENT". If I put this agreement in with the Chess Fighter download is that sufficient? Yes, it should be enough. Lc0 had also the problem from the opposite side, GPL by default doesn't allow bundling closed so...

I think that you would prefer to call the policy network move an N node search, where N is the number of legal moves, since all of these moves have been given a probability. But the actual node reached by each move has not been visited, so technically I think they are correct. Policy head performs ...