TalkChess.com

You should be able to use the 32930-112x9-se distilled net for cpu tournaments.

dkappe wrote: ↑

Tue Mar 26, 2019 9:47 pm

Can’t get it to build on Linux. I did post the source to github:

https://github.com/dkappe/ThothFish

I didn’t notice the .o and exe files. I removed them. Now builds like a charm.

Can’t get it to build on Linux. I did post the source to github:

https://github.com/dkappe/ThothFish

At the usual place: https://github.com/dkappe/leela-chess-w ... d-Networks

Are you going to swap in the new version for tcec? Maybe it can overtake the spaghetti code of lc0?

It is using mcts search in tcec. The alphabeta version has not been released yet. And the engine playing in TCEC is officially “Allie+Stein” as it is using Mark’s excellent SL “Stein” network. Without that, Allie could not play chess. Just want to note, that it's not correct to say that "Stein" is ...

Excellent news. The more the merrier.

Leo wrote: ↑

Mon Mar 11, 2019 4:52 pm

What does distilled mean?

See the page. It contains a description.

https://github.com/dkappe/leela-chess-w ... d-Networks

I'm done with 11258. I have a full size net cooking right now, but will distill a few 32930 nets after.

Just to be clear, I have no issues with lc0 or Allie. My issue is with the somewhat arbitrary and vague TCEC rules governing Neural Network engine participation. These rules state: Definition A neural network is a computer system modeled on the human brain and nervous system. For the purpose of TCEC...