Search found 765 matches

by AndrewGrant
Thu Nov 19, 2020 9:03 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: You've trained a brilliant NN(UE) King-Piece Network. Now what?
Replies: 22
Views: 1388

Re: You've trained a brilliant NN(UE) King-Piece Network. Now what?

The other thing to remember is that this technology (NN on CPU) is still relatively new, having only been introduced to the wider computer chess community a few months ago from the shogi world, so I did not expect there to be many papers on this subject in the first place not in the Japanese langua...
by AndrewGrant
Thu Nov 19, 2020 8:55 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: You've trained a brilliant NN(UE) King-Piece Network. Now what?
Replies: 22
Views: 1388

Re: You've trained a brilliant NN(UE) King-Piece Network. Now what?

smatovic wrote:
Thu Nov 19, 2020 8:53 am
Dude, you are funny, "Don't copy SF NNUE, you are lame!!!111", "Where are the papers for NN implementations???", "I got +100 Elo sitting in here but won't publish it!!!111", what shall 'we' make out of this?

Maybe you should go commercial? Seriously, could relax some things?
?
by AndrewGrant
Thu Nov 19, 2020 7:57 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: You've trained a brilliant NN(UE) King-Piece Network. Now what?
Replies: 22
Views: 1388

You've trained a brilliant NN(UE) King-Piece Network. Now what?

Skip to the bottom if you don't want to humor me.... Monologue :::: So I wrote an NNTraining tool in C, made it very well threaded, cranked up optimizations, wrote AVX2 code for portions to perform weight updates faster, came up with a novel addition to Adam to limit throughput when dealing with spa...
by AndrewGrant
Wed Nov 18, 2020 10:38 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Pytorch NNUE training
Replies: 58
Views: 4814

Re: Pytorch NNUE training

Gary, I'm trying to output my networks with an SF format. You have a couple different versions going in in your repo, but do https://github.com/glinscott/nnue-pytorch/blob/master/serialize.py and https://github.com/glinscott/nnue-pytorch/blob/master/model.py correspond to what would be matching Nodc...
by AndrewGrant
Wed Nov 18, 2020 1:56 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: 4 x Intel Xeon good idea?
Replies: 23
Views: 1101

Re: 4 x Intel Xeon good idea?

If running parallel single-core games, the total nps is a good indication of the performance, but multi-threading is lossy. At equal overall nps, more threads plays weaker. The scaling efficiency from 12 (24) to 48 (96) threads is really poor, so matching or moderately beating the modern 12 cores i...
by AndrewGrant
Tue Nov 17, 2020 11:03 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: Only +150 Elo (Halogen 8.1 test) ?
Replies: 4
Views: 420

Re: Only +150 Elo (Halogen 8.1 test) ?

Halogen 8.1 has for sure over 3100 Elo points. Let's wait for the final result. Hard days for Rhetoric 1.4.3 x64. Only the begining :) This was committed _after_ 8.1's release :) https://github.com/KierenP/Halogen/commit/f881c3aecb3b54fc495a20949ef1252eae8f857e Fast tracking to TCEC :) Hello Andrew...
by AndrewGrant
Tue Nov 17, 2020 8:58 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: 4 x Intel Xeon good idea?
Replies: 23
Views: 1101

Re: 4 x Intel Xeon good idea?

If you are trying to buy an existing server at a discount, Intel will probably have something good for you. If you want the best possible performance per dollar right now though, AMD wins the top 10 or so possible options. A good source for data is : http://ipmanchess.yolasite.com/amd---intel-chess-...
by AndrewGrant
Tue Nov 17, 2020 8:17 pm
Forum: Computer Chess Club: Tournaments and Matches
Topic: Only +150 Elo (Halogen 8.1 test) ?
Replies: 4
Views: 420

Re: Only +150 Elo (Halogen 8.1 test) ?

Sylwy wrote:
Tue Nov 17, 2020 8:03 pm
Halogen 8.1 has for sure over 3100 Elo points. Let's wait for the final result. Hard days for Rhetoric 1.4.3 x64.
Only the begining :) This was committed _after_ 8.1's release :)
https://github.com/KierenP/Halogen/comm ... 2eae8f857e
Fast tracking to TCEC :)
by AndrewGrant
Tue Nov 17, 2020 3:35 am
Forum: Computer Chess Club: General Topics
Topic: Maybe not the best diversity of strongest chess engines under development
Replies: 12
Views: 1433

Re: Maybe not the best diversity of strongest chess engines under development

Unfortunate, and entirely unsurprising in regards to the NNUE similarities. Perhaps, some hope though. Even though Komodo and Stockfish are, it appears, trained on the same code base, the differences between their evals, and its usage in training, is enough to have at least some diversity. You don't...
by AndrewGrant
Mon Nov 16, 2020 11:49 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Pytorch NNUE training
Replies: 58
Views: 4814

Re: Pytorch NNUE training

Yes, this comes from using the GPU (hooray pytorch!). As well, Sopel implemented a super fast C++ data parser, which feeds the inputs to pytorch as sparse tensors, which was a very large speedup (since the inputs to the first layer are super sparse). What was the batch size for this? That makes a b...