Search found 242 matches

by Pio
Fri Jan 15, 2021 2:49 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

I thought about that this morning too, that it could explain it as well. Another problem might be that it looks like you haven’t computed the K from your test data (see code below). That means that the mapping between centipawn scores and probabilities might be completely off. Calculate the K Value...
by Pio
Fri Jan 15, 2021 1:09 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

... BTW, are you sure the results in your data set are given from side-to-move POV, and not from white POV? Somehow this were not there when i quoted it. http://talkchess.com/forum3/viewtopic.php?f=7&t=74221&p=847806&hilit=epd+ccrl#p847806 ... 3. Marks each EPD with the PGN result (0.0, 0,5 or 1.0 ...
by Pio
Thu Jan 14, 2021 7:52 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

Probably the balancing should be like 1/3 wins, 1/3 draws and 1/3 losses. If you don’t balance the games with respect to results it will try to predict that almost everything is a draw. Will you go sailing only during winter if you know that 90 % of the drowning accidents occur during summer? Puttin...
by Pio
Thu Jan 14, 2021 7:27 am
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

That the numbers differ between the experiment of step size 5 and 1 can then be explained with that it is much more important to get a fine grained eval than a good predictor of wins/losses. Did you noticed that the total error with stepsize 5 was less than with 1? significantly! The total error fo...
by Pio
Wed Jan 13, 2021 11:43 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

But we talk about why the tuner will diverge the mg an eg value while keeping the average of both constant. That much I understood. But doing this should greatly drive up the rms error, agreed? Because keeping the average constant will prevent any effect on the prediction for those positions that h...
by Pio
Wed Jan 13, 2021 11:21 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

But we talk about why the tuner will diverge the mg an eg value while keeping the average of both constant. That much I understood. But doing this should greatly drive up the rms error, agreed? Because keeping the average constant will prevent any effect on the prediction for those positions that h...
by Pio
Wed Jan 13, 2021 11:20 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Tapered Evaluation and MSE (Texel Tuning)
Replies: 89
Views: 1914

Re: Tapered Evaluation and MSE (Texel Tuning)

But we talk about why the tuner will diverge the mg an eg value while keeping the average of both constant. That much I understood. But doing this should greatly drive up the rms error, agreed? Because keeping the average constant will prevent any effect on the prediction for those positions that h...
by Pio
Tue Jan 12, 2021 11:03 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Why using the game result instead of evaluation scores
Replies: 11
Views: 281

Re: Why using the game result instead of evaluation scores

The main point is that with this training you would learn your evaluation to mimic a certain engine. That would also make it copy the errors of that engine. There is no arguing with game results, though. Even engines that misevaluate a dead draw as a win will eventually draw, when they have to fini...
by Pio
Tue Jan 12, 2021 4:39 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Training data
Replies: 14
Views: 370

Re: Training data

Training data needs to be of low quality at first. The positions should be bad that your current evaluation weights can see the difference of how bad it is, that way it can adjust its weight. Then a much improve training data but still bad so that the engine has the opportunity to change the weight...
by Pio
Tue Jan 12, 2021 2:45 pm
Forum: Computer Chess Club: Programming and Technical Discussions
Topic: Training data
Replies: 14
Views: 370

Re: Training data

Due to my current experiences, I realized again how crucial it is for the tuning is to have useful data. I have read some posts about this in the past. Nevertheless, I would like to discuss a handful of questions on this topic here. 1. source of the data 1a. A quick way to get data is to use public...