Algorithm Optimisation Championship. - page 112

New comment
 
Event:

First, you willquote "Mr Zelinsky' s"provocations and trolling.
If you want to find quotes of provocation from Mr Zielinski, go to the links he puts up saying the topicstarter is insulting him, but look a little higher.
 
Реter Konow:
I don't get it, and where is the provocation and insult?
Don't attribute to me what I didn't write
First youcite the provocation and trolling of "Mr Zielinski"
 
Реter Konow:
If you want to find quotes of provocation from Mr Zielinski, go to the links he puts up saying the topicstarter is insulting him, but look a little higher.

You'll be telling the moderators that later, along with Dick.

Go on with your championship!!!

We're all waiting....

 
Yuriy Asaulenko:
Actually, it's all together and it's called the interface. I agree that it's not kamilfo without it.

The connection interface was shown a dozen other pages ago.

 
Реter Konow:

4. Your endless participation in the process of your endless insult by the topicstarter (as you claim) is surprising. Why are you present where you are insulted and disrespected?

Firstly, Dick is not just being rude and insulting to me (a key and important point that for some reason is being overlooked). There is no need to demonise my statements and my presence on a level playing field.

Secondly, I may be sadistic by nature and I find it interesting to watch Dick's intemperance (even more accurately, rampaging in insults and boorishness).

As for respect -- it's impossible to please everyone -- being undesirable (there's even a special word, odious) to part of the community is okay.

The ideal person -- is a mentally ill person -- who, apart from his illness, by definition has no faults.

 
Реter Konow:

I want to create an algorithm.

To do that, I need an FF library that includes the entire set of exportable functions, including the FF and the connection interface. The connection interface you have already laid out.

The FF, as you yourself know, is not an analytical function, but a function which includes an analytical function in which an array with parameter values is substituted.

Without the FF and the entire library, I cannot create an algorithm to "train" and compete.

Given that you yourself know it all, and I know that you know it all, your advice to "train" sounds a bit ridiculous.

The connection interface is already shown a dozen other pages ago. Note, when Yuri asked to check the minimum of his function, I simply pasted the function into the ff.ex5 library and found the minimum. I didn't have to write an algorithm to solve Yuri's problem. I don't have to rewrite the algorithm for any problem. Do you know why? - Because I have it. You should have it by now, too. But I don't.

Thank you for your support. But I can handle myself.

 
Andrey F. Zelinsky:

Firstly, Dick is not only rude and insulting to me (a key and important point that for some reason has been overlooked). There is no need to demonise my statements and my presence on a level playing field.

Secondly, I may be sadistic by nature and I find it interesting to watch Dick's intemperance (even more accurately, rampaging in insults and boorishness).

As for respect -- it's impossible to please everyone -- being undesirable to part of the community -- that's fine.

The ideal person -- is a mentally ill person -- who, apart from his illness, by definition has no faults.

If you are a sadist by nature, you are also an artist because you skilfully create a masochist.

I personally do not demonize you and treat you adequately. Your presence does not bother me. I would even say on the contrary.

The society is diverse enough within itself and some part of it applauds you.

If that is the basis of your motivation to be present, you are obviously not qualified to actually compete in the championship.

If you're not a competitor, you're a spectator who occasionally runs out on stage to show the rest of us your "PA". Please accept my applause as well.

 
Andrey Dik:

The connection interface is already shown a dozen other pages ago. Note, when Yuri asked me to check the minimum of his function, I simply inserted the function into the ff.ex5 library and found the minimum. I didn't have to write an algorithm to solve Yuri's problem. I don't have to rewrite the algorithm for any problem. Do you know why? - Because I have it. You should have it by now, too. But I don't.

Thank you for your support. But I can handle myself.

Especially for you, I'm not providing any support. It's just my role here.

I see where you're going with this. I have to create an algorithm without seeing the FF library and the calling script.

I'll think about how to do it, but I hope you won't "tighten the screws" so much that you won't leave me time to work on the algorithm.

 
Реter Konow:

1. Well, if you're a sadist by nature, you're also an artist, as you skilfully construct yourself as a masochist.

...

2. If that's the basis of your motivation to attend, you're obviously not qualified enough in that area to actually compete in a championship.

...

1. If I'm a sadist, I'm also a masochist -- that's a simple conclusion -- you as an expert in "optimization" should have made it without assumptions.

2. If you read my posts back in the first thread -- notice -- it's literally my 3-4th post out of 130 already written by someone (it's about like-"common", that I do trolling) --https://www.mql5.com/ru/forum/84457/page13#comment_2495423-- I expressed my attitude to "championship" - why participate in absurdity - because how it ended - we clearly see after a while.

p.s. I will even move this post here:

Forum on trading, automated trading systems and testing trading strategies

Do you have your own developments of evolutionary algorithms?

Andrey F. Zelinsky, 2016.05.15 11:25

It's clearly obvious that it's too early for a contest. The topic is little known and few people use it in their practice.

For the contest to be interesting and useful:

1. it should be run by an MC in the same manner as similar competitions have been in the past, for example on panels

2. provide a series of publications, articles, code base, thematic threads -- where, using examples of theory and practice, EA should be explained how to use it in practice and in trading

And when the active part of the forum (at least ten participants) will be ready, then we will hold the contest.

At the moment:

a) there are no spectators who understand what we are talking about

b) there is no jury, that the community would understand that the jury is competent

c) there are no participants who understand what we are talking about

Bottom line:

-- if a contest is held now -- it will be a contest between a couple of contestants who will provide an incomprehensible and unambiguously shitty result and will have to pick a winner out of nothing


 
Реter Konow:

...

In the championship FF the function (the second one in the picture I have shown)

// max = 8.10621358944233528 [-5.0; 5.0] 
double ff1 (double x, double y)
{
  return(0.3 e1 * (double)(int)pow ((double)(1 - x), (double)2) * exp ((double)(-x * x - (int)pow ((double)(y + 1), (double)2))) - 0.10 e2 * (0.2 e0 * (double)x - (double)(int)pow ((double)x, (double)3) - (double)(int)pow ((double)y, (double)5)) * exp ((double)(-x * x - y * y)) - exp ((double)(-(int)pow ((double)(x + 1), (double)2) - y * y)) / 0.3 e1); 
}

is duplicated 250 times. For each copy of this function a peak offset is given according to the referee generated values.

Thus, the maximum of the total FF is known as 8.10621358944233528*250=2026.55339736058

The only thing I will add is the offset, so that the maximum is not a fractional number for ease of visual control.

1...105106107108109110111112113114115116117118119...133
New comment