Search found 403 matches

by brianr
Fri May 22, 2020 10:44 pm
Forum: Computer Chess Club: General Topics
Topic: Leela - Dance of the knights
Replies: 16
Views: 1120

Re: Leela - Dance of the knights

Just ask on the Leela chess Discord in the #help channel.

https://discord.com/channels/4254194825 ... 1713154049
by brianr
Fri May 22, 2020 10:38 pm
Forum: Computer Chess Club: General Topics
Topic: Leela - Dance of the knights
Replies: 16
Views: 1120

Re: Leela - Dance of the knights

Lc0 and most of the other NN engines should be run with 2 threads, not 1 (or number-of-GPUs + 1). As far as I know, it's not important, one core is enough. The calculation of the NN engines is done almost exclusively via the GPU. BTW, a very strong Nvidia-RTX2080 was used here... Exactly 2 threads ...
by brianr
Fri May 22, 2020 8:02 pm
Forum: Computer Chess Club: General Topics
Topic: Leela - Dance of the knights
Replies: 16
Views: 1120

Re: Leela - Dance of the knights

Lc0 and most of the other NN engines should be run with 2 threads, not 1 (or number-of-GPUs + 1).
Incidentally, more than that actually hurts performance (very unlike A/B engines).
by brianr
Tue May 05, 2020 12:47 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: AlphaZero
Replies: 19
Views: 2974

Re: AlphaZero

A couple of things. The CCRL Standard Dataset is quite valuable as it provides a trained net benchmark to compare against a net you have trained from the same data. However, as mentioned earlier, the Bad Gyal data will produce stronger nets having more Policy information. And, of course, the actual ...
by brianr
Thu Apr 30, 2020 2:11 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: AlphaZero
Replies: 19
Views: 2974

Re: AlphaZero

Don't know what you mean by pre-training, but a suggestion. Doing supervised learning (SL) from games with only the move actually made set to 1 turns out to reduce net strength by about 100-150 Elo. Training games from Lc0 include policy probabilities for all moves and produce stronger nets. It is p...
by brianr
Mon Apr 27, 2020 7:07 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: AlphaZero
Replies: 19
Views: 2974

Re: AlphaZero

Sometimes the answer is simply because that's what Alpha Zero did...

Experiment and see what works for you after you have established a working baseline.

Have fun.
by brianr
Sun Apr 26, 2020 11:04 pm
Forum: Computer Chess Club: General Topics
Topic: Sapeli 1.0 - New chess engine
Replies: 137
Views: 43193

Re: Sapeli 1.0 - New chess engine

Move ordering and pruning is getting to the heart of the individual ecosystem that each engine has, so what works in one may not help at all in another. I've just never found anything other than scanning the current move list subset and picking the highest value to be any improvement. The first move...
by brianr
Sun Apr 26, 2020 8:27 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: AlphaZero
Replies: 19
Views: 2974

Re: AlphaZero

Suggest reviewing this first:

https://github.com/Zeta36/chess-alpha-zero

or here for a simpler game like Connect4:

https://github.com/suragnair/alpha-zero-general
by brianr
Sun Apr 26, 2020 8:25 pm
Forum: Computer Chess Club: General Topics
Topic: Sapeli 1.0 - New chess engine
Replies: 137
Views: 43193

Re: Sapeli 1.0 - New chess engine

the general thinking was that "fancy" sorting does not pay as so much of the tree is pruned in the first place That holds only for the "pick the best of the remaining list and put it to the current place" sort, but not compared to sorting the whole move list up-front with an n^2 algo. All the moves...