Search found 4288 matches

by Ferdy
Wed Oct 28, 2020 3:35 am
Forum: Computer Chess Club: General Topics
Topic: A better rating method ?
Replies: 10
Views: 473

Re: A better rating method ?

The naive assumption that is used for standard elo calculation has been shown many times to not capture enough of an engine's properties. This thread is about discussing how could be designed a system that capture more of those properties. The normal method to measure engine's strength is by Win/Lo...
by Ferdy
Sat Oct 24, 2020 8:03 pm
Forum: Computer Chess Club: General Topics
Topic: Best way to teach humans some opening reprtoire
Replies: 3
Views: 248

Re: Best way to teach humans some opening reprtoire

Bookup is good indeed. You open your old books then input in bookup then analyze and backsolve. :D

For tabs and phones this one is good. There is training with random and sequential, speed test and other options like which positions to train (low scores, untrained, schedule).
by Ferdy
Sat Oct 24, 2020 7:44 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: delete informant symbols before a given move or ply nr in big pgn database
Replies: 7
Views: 420

Re: delete informant symbols before a given move or ply nr in big pgn database

Uploaded a python script using python-chess lib. * Install python 3.8 or newer. * Install python-chess lib * Open command prompt * type pip install python-chess * Download the repo, or pc_0001.py file Usage: python pc_0001.py --input in.pgn --no-nag-ply 10 The output will be saved in out_in.pgn type...
by Ferdy
Sat Oct 24, 2020 5:11 pm
Forum: Computer Chess Club: General Topics
Topic: Best way to teach humans some opening reprtoire
Replies: 3
Views: 248

Re: Best way to teach humans some opening reprtoire

I think chessable is doing spaced repetition. Learning would be better if during learning humans would give feedback to the positions, i.e. move is easy, normal, or difficult to find. This way the program would train more often on those positions that are difficult. A human feedback can be automatic...
by Ferdy
Tue Oct 20, 2020 1:33 pm
Forum: Computer Chess Club: General Topics
Topic: CTF Survey - Finished. Results in final post.
Replies: 74
Views: 5929

Re: CTF Survey

2. CTF should be closed
by Ferdy
Sun Oct 11, 2020 9:32 pm
Forum: Computer Chess Club: General Topics
Topic: INTERESTING Engines with Unique Styles (Unlike SF)
Replies: 56
Views: 4279

Re: INTERESTING Engines with Unique Styles (Unlike SF)

This is a fresh thread for Thorsten's idea of making a list of all "interesting" engines. For me, in no particular order, they are: Teki 2 Thinker 5.1c Passive Baron 1.8.1 Gandalf 6 WChess 1.06 Vajolet 2.03 ProDeo 1.0-1.6 (assuming no personality used) ZuriChess Graubuenden Alfil 8.11 Stash 21.0 (n...
by Ferdy
Sun Oct 11, 2020 8:46 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Is there a fen suite to ...
Replies: 6
Views: 626

Re: Is there a fen suite to ...

Is this pure wood count, or do you take material imbalance into consideration also? Pure wood count. 100,300,300,500,900 Uploaded material epds in github. pawns.epd r4rk1/p2b2pp/1q2pb2/3p3Q/3p4/3B4/PP3PPP/R1B1R1K1 b - - ce 100; r7/p2q1rk1/2p4p/4P1p1/1BQ1Ppb1/1P6/P1P3PP/4RRK1 w - - ce 200; ce values...
by Ferdy
Sun Oct 11, 2020 6:31 am
Forum: Computer Chess Club: General Topics
Topic: Yet another parameter tuner using optuna framework
Replies: 36
Views: 4136

Re: Yet another parameter tuner using optuna framework

No4b wrote:
Sat Oct 10, 2020 1:58 pm
I tried to use this tuner for my Drofa engine with some nice results.
Thanks for info. Looking forward for your next test.

In case you are not aware, I run a fixed depth test comparing different algorithms. Would like to do a test with time control next.
by Ferdy
Fri Oct 09, 2020 7:39 pm
Forum: Computer Chess Club: General Topics
Topic: Determining evaluation by a genetic algorithm
Replies: 5
Views: 438

Re: Determining evaluation by a genetic algorithm

We had a fun experiment on GA with stegemma.

There are also optimizers at nevergrad on GA.
by Ferdy
Mon Oct 05, 2020 11:34 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: How to calculate piece weights with logistic regression?
Replies: 12
Views: 2091

Re: How to calculate piece weights with logistic regression?

Put up a repo on github for piece value estimator . It uses different linear models from sklearn lib. Sample output fen P-p ... Q-q result 0 r3r3/pbp1kp2/2p2bpp/2N1p3/8/5B2/PPP2PPP/2KRR3 ... -1 ... 0 1 1 3rr1k1/pb4pp/2p1n3/1p1n4/1P1qN3/P7/B1Q2PPP/3R1... 0 ... 0 0 2 6k1/p1p3p1/1p5p/4p3/2P2n2/2P2B2/P4...