75,000 options - 4GB RAM and 4GB disk cache not enough??? - page 4
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
GO is orders of magnitude faster,
There is just a glitch in the software.
For genetic optimisation in any memory parameter space, there is practically no memory required.
Same thing, only instead of 21600 I wrote 34440 runs, but it uses 10 times more memory - for 215 runs (8m22sec) 494Mb. Are you and I the first to see this?
By the way, you still haven't posted the full code of the Expert Advisor here (it's not your code). I don't ask for the full data for test replication for nothing, but you are still unwilling to provide it and focus on asking questions instead of getting the exact answer.
By the way, you still haven't posted the complete code of the Expert Advisor here (it's not your code). I don't ask for the full data for test replication for nothing, but you are still unwilling to provide it and focus on asking questions instead of getting an exact answer.
Now I don't understand something anymore.
Yes, it's not my code at all. Did I write somewhere that it's mine? It's published in the public domain and it's copyrighted.
The man made an open template and distributed it to everyone.
I ran the tests we're discussing on it with exactly those parameters that were inserted several posts ago.
I cannot test my code in the tester with such memory and resource consumption unambiguously.
Although the passes of the tester are about 50 times less, my code is what I have sent you ^2 (squared), i.e. one by all results of the other. let's deal with it without the square. or didn't you get the mail? I am posting it here now. I hope to finish the optimization directly in my EA.
How much memory do you need? 40 Gbytes? What for? Then 400? The data I test using in a text file take 1.5 mb. In binary form it is at least 5 times less, i.e. 300 kb. The reasonable maximum for the tester's history is 300 kb in ^2 - about 90 mb.
The point is that a perfectly good and stable program written in a low level language other than Java or Net at the moment has a vulnerability and becomes unstable. the program outperforms metastock, vels, omega and god knows what else in terms of stability, speed and portability.
fine it needs 40gb, but i can't just insert chips on the fly. at least let it write these gigabytes somewhere instead of crashing out without saying goodbye.
furthermore, you can ask me directly where I'm aiming and not claim it for me.
You have to ask a question to get an answer.
Questions on the previous 2 pages I asked 10 thousand. no answers did not get. no problem, I understand that the customer (if any) - docs, and I'm to you a virtual nickname that has "surfaced" and most likely will now dive into the same place where they resurfaced.
p.s. if you're not happy with the tone or content please tell me directly how it should be here.
Alexey
tel 89262090000
2090000.com
I will do so today and publish the results.
Thanks for the code provided.
They've posted an updated build 197: they've "trimmed" the actual memory allocation in severe cases. But there is no way to pull out 57 billion.
sane, download the updated 197 build and try again, please.
In genetics, scrolling through NN billion passes of the coverage area is not a problem. The problem is in multiple visualization of results and available memory.
In any case - parameter overclocking for tens of billions has nothing to do with real optimization tasks. Our task is to perform our work with full calculations and obligatory tabular and graphical visualization, so that anyone can visually perceive the results and access any run with a mouse, for normal tasks (search area up to 2 billion variants).
Demonstrate the results of your genetic tester run right here with the same parameters you specified on the simple MACD Sample. I'm sure it won't be difficult for you to translate the code into Easy Language and show your results.
There are many ways to solve the problems you describe (actually, there are no problems at all).
But it makes no sense to talk about them here in the forum, especially since you immediately take the stand.
Well, you do not want to make your product better, I then why . ?
"Overclocking parameters by tens of billions has nothing to do with real optimization tasks".
Strange, but men do not know ... and solve all sorts of optimization problems in the space of 10^100...0.
For example Neural Networks.
To train neural networks means to optimize function with thousands and tens of thousands of real parameters (weights on neuron inputs).
to carry out its work with full calculations and obligatory tabular and graphical visualization
And what is the problem with visualization?
Is it about storing a multidimensional sparse matrix?
If so, there are many different methods of compactly representing such matrices in memory.
(Including with very fast cell access).
Demonstrate the results of running your genetic tester right here with the same parameters as you indicated on the simplest MACD Sample. I am sure it won't be difficult for you to translate the code into Easy Language and show your results.
Why and what to show?
Why do you need it? (Don't you believe me?)
Why do I need it, to spend half a weekend doing meaningless work for me?
(And given the explanation of the results, and say you want to see the results on EURUSD H1, and I'll have to find somewhere to get them for Omega, it may take me more than one day).
Believe me, I ran the optimizer on a system with several hundreds of real parameters (600 - 800) to test it.
Our customers sometimes optimize several dozens of parameters.
By the way, our optimization criteria are not "final balance" (or rather, the criterion is calculated by the user).
But the version we are developing now will contain optimization by many criteria at once.
But what does all this have to do with the subject at hand?
________________________________________________________
PS.
1. All the above is simply my private opinion.
I just think so, and I'm not going to waste my time on pointless evidence ...
2. My only purpose in this discussion was to help you make MT even better and remove unnecessary restrictions.
But if you don't want that yourself, I don't need it all the more.
3. If you don't want to call it a glitch, let's consider it a feature ...
I don't want to quarrel with you, let's consider this topic closed.
When a person poses questions, and even with a bias towards theorizing, the answers are only one - "let's move on to practice and real calculations".
Please only communicate with clear practical links.
Many critics respond to the question "give practical evidence" in this way:
Why and what to show?
The last time this happened was with mandor, who only paid lip service to statements.Why do you need it? (Don't you believe me?)
Why do I need it, to spend half a weekend doing work that makes no sense to me?
You, by the way, have already been banned precisely for failing to provide proof of your words.
Clearly prove the practical correctness of your words. But not in words, on full display with proofs. We do our work openly and publicly defend our ideas, we spend effort. You should try to do the same.
be very specific,
what you want me to say.
Proof of WHAT,
and in WHAT form.
Renat,
be very specific,
what you want me to say.
Proof of WHAT,
and in HOW.
As I asked earlier:
Demonstrate the results of your genetic tester run right here with the same parameters as you specified in the simple MACD Sample. I'm sure it won't be difficult for you to translate the code into Easy Language and show your results.
After that you can continue the discussion. But continue exactly after publishing your results.My optimiser is not like yours.
And what do you want to see there?
The fact of the run itself?
In what form should I quote the results of the run?
My optimiser is not like yours.
And what do you want to see there?
The fact of the run itself?
I didn't say for nothing "some people count in their head and some people show everything".