Search found 5 matches

by castlehaven
Mon Nov 30, 2020 10:56 pm
Forum: Computer Chess Club: General Topics
Topic: TalkChess Is Back
Replies: 20
Views: 5545

Re: TalkChess Is Back

As a relatively new member of this community, I am writing now to express my appreciation for the very useful discussions and information that I have found here. I did not realize how much I was relying on it until it was gone for a few days. I know there has been a lot of drama around here in the l...
by castlehaven
Tue Nov 17, 2020 2:12 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Pytorch NNUE training
Replies: 61
Views: 21835

Re: Pytorch NNUE training

Are we all using “epoch” the same way in this discussion? In the nodchip trainer, the time to train depends (first approximation) on the total number of positions. In this case 256M positions in ten minutes is darn fast, no matter how we break up the batches. The original nodchip syntax would define...
by castlehaven
Sun Nov 15, 2020 6:56 am
Forum: Computer Chess Club: General Topics
Topic: CoronaVirusChess
Replies: 67
Views: 16591

Re: 14th ACCA World Computer Rapid Chess Championships

FWIW, I read Albert’s comment completely as a political joke about crazy Trump COVID cures. Such jokes (about sunlight, bleach etc.) are ubiquitous these days in the USA. Everyone works so hard on their engines here, so we are all primed for attacks. I understand that. But this really looked like a ...
by castlehaven
Thu Nov 12, 2020 5:28 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tensorflow NNUE training
Replies: 7
Views: 2087

Re: Tensorflow NNUE training

I have read some of the tensorflow vs. pytorch debates but come out still feeling confused. Is there any reason to prefer one of these environments to the other for NNUE development? Or is it 99% driven by whichever one you are most familiar with? I ask because I am looking to starting from scratch,...
by castlehaven
Sat Oct 17, 2020 3:07 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: BONA_PIECE_ZERO
Replies: 18
Views: 4461

Re: BONA_PIECE_ZERO

Ah, it seems I had misunderstood the bona-piece-zero. I had thought it was a clever way to encode that the previous move had been a capture, thus allowing the NN to see that the position might require a recapture if it currently had a material imbalance. Since that is not the case, I wonder how the ...