Search found 360 matches

by noobpwnftw
Fri Aug 02, 2019 3:28 pm
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

How was the "winrate" for each position determined? I suspect it is logistic function based on score. The score is from "the scores been back-propagated using a weighted averaging function" see first post of this thread. In javascipt when winrate is missing (by design) for mate scores, I use // 201...
by noobpwnftw
Fri Aug 02, 2019 2:31 pm
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

My API endpoint supports HTTPS, so you could just change the base URL and get rid of the unsafe script warnings.
by noobpwnftw
Thu Aug 01, 2019 5:28 am
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

Ferdy wrote:
Wed Jul 31, 2019 12:50 pm
How do I know if positions from the run are saved in cloud?
It will be there, just currently the DB has a very long queue, need a while to process them.
by noobpwnftw
Wed Jul 31, 2019 7:48 am
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

finally there exists a "Discover" program people can run, which loads any UCI-compatible engine and play games, contributing moves encountered during the process. I tried to compile ChessDBDiscover and it seems to work. Created an engine dir and put engine.exe, a renamed stockfish dev engine, then ...
by noobpwnftw
Wed Jul 31, 2019 7:45 am
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

the back-propagation algorithm came up with the scores in the first place, one shouldn't try to put chickens back into eggs But if 10...Qc7 eventually leads to a score of 0.00 while 10...O-O leads to a score of -0.05 and 10...Nb7 leads to a score of -0.02, against white's best play, surely Qc7 shou...
by noobpwnftw
Wed Jul 31, 2019 3:10 am
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

My scores are evaluations of moves at current position, in relation to those of other moves of the same position, they have no meaning to any subsequent positions their corresponding moves may lead to(the back-propagation algorithm came up with the scores in the first place, one shouldn't try to put...
by noobpwnftw
Wed Jul 31, 2019 12:43 am
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

Ferdy wrote:
Tue Jul 30, 2019 2:45 pm
See there is no 'winrate'. Its winrate should have been 0.0 assuming winrate has a range [0, 100], %.
There is no winrate field for mate scores or exact TB scores(not yet available). It is by design.
by noobpwnftw
Mon Jul 29, 2019 8:11 pm
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

Very interesting, thank you very much. One question: When I find a leaf in the tree (position after last given move not found) does it mean that this was evaluated with Stockfish depth 22 and no other analysis starting from that position was performed? Yes. As an anecdote: I played a Lichess blitz ...
by noobpwnftw
Mon Jul 29, 2019 1:44 am
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

Nice work, Noob. Sounds like you've got 35-40 million games in that database. Any word on the composition, i.e. % human, % engine? This work would be greatly improved with a custom GUI that included your evaluation data, the empirical WLD data (flawed though it may be) and information on who played...
by noobpwnftw
Sun Jul 28, 2019 4:35 pm
Forum: Computer Chess Club: General Topics
Topic: Database snapshot
Replies: 114
Views: 14829

Re: Database snapshot

My weighted averaging function also factors in the number of good reply moves so that it can tell the difference between different 0.00 lines and apply penalties accordingly, also gradually degrade sharp scores to prevent the back-propagation becoming fully min-maxed. Regardless of how you want to a...