Machine learning in trading: theory, models, practice and algo-trading - page 1692
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
Ok, got it. It turns out that if it works well with dropout greater than 0.5 then there is a lot of unnecessary stuff in vector.
It's kind of... It's not that it's superfluous, it's just that she starts to remember, not generalize. And here the dropout at each epoch cut off half of her neurons, it forces her to sort of parallelize information on all neurons (generalize) because it is not clear which neurons will be cut off next time, and it is necessary to learn further.
I'm not arguing, I'm the instigator, so to speak.
Well, what's the point of arguing? You will not share your honestly earned profit with me? And I do not promise to compensate for your losses
))))
As for the banks, they have different objectives, but I can tell you that banks lose money too, and regularly ;)
As for the brokers, the real dealers are engaged in training, it really works. - i think that the objectives are different too.
SZS: i remember the fight with heretics, now it's time to find out who is right and who is to be burned at the stake ))))
UPD: it's time for great stories....
Here I have also made a chart, EA generates TS, TS is not perfect, but imho, this chart may be used for further work
I don't think so, it's not ideal, but it's possible to go further with this chart.
Because on the real you will increase the risk, which makes TS-ka red rag inevitably leading to the drain.
Igor, the problem is that martin in the tester and on the real are not the same thing.
The reason is that in real time you'll increase risk, which makes the TS a red rag inevitably leading to failure.
fuh....
i think i already wrote, i asked about position management, the answer was only martin-martin
i've already done it, i've already done it with an asctrend indicator, i've already done it with an asctrend indicator - it's a test and a forward - now it's a real position management
here it is - test + forward - now it is based on real ticks
and the point of this chart is not in the beautiful balance chart but in the automatic search for these wonders, fuck!
SZZY: forum communication is something, any question will get an answer, but unfortunately the answer will be prepared in advance ))))
But in the meantime, EA found a way to work with orders, stopped the optimization
But to my question why some EAs pass the forward testing, while others do not, and how to determine this, .... That's all the answers: "martin f*ck."
fuh....
Is there a report there with numbers... please?
This is the kind of stuff that gets thrown around:#32Is there a report with numbers?
Yes, there is.
I stopped the optimization, I will send this tester report to the PM, you can analyze it
Optimization 2018.01.01 - 2019.05.01, then test with forward to today - screenshot above, test by real ticks
If you don't know what i'll do if i don't do it, i'll try to win in the end... I think I'll find more information
yes there is
And the reports can be generated several times an hour, I'm writing openly that the search for TS can be automated, and I'm asking how not to leak even if the TS passed the forward test
i don't know, i'm not sure, but i'll try to find the answer. i think i`ll find more information
fuh....
i think i already wrote, i asked about position management, the answer was only martin-martin
where's the martin? my chart is a counter closure by limit, if CHO - so the tester found the optimal order system...
If it will be positive on forwards (ASR>2), then optimize separately MM and execution, all together optimize - the way to the downside. In general optimization is a lamer thing, but since...
No, this will not work, at least optimize the system without MM first, and if it is in the formard will be somehow in the plus (ASR>2), then optimize separately mm and execution, all together optimize - the way to the flop. In general optimization is a lamer thing, but since...
I do not develop TS for hedge funds and banks, maybe your tasks are different, this may be a misunderstanding
let's reduce our conversation to a well-established question - do you have a state?
I am in the process of research, you, as I understand it all works and can you share your real experience?
I don't think it's necessary to teach me how to test the TS, I've been studying this topic for a long time. And to explain for the 101st time that you call it MM, this is your generalization, and you don't even realize that to close the profit in time is as complicated as not letting the losses to run out - I just can't tell it.
I do not develop TC for hedge funds and banks, perhaps your tasks are different, this may be a misunderstanding
Let's reduce our conversation to a well-established question - do you have the state?
I am in the process of research, you, as I understand it all works and can you share your real experience?
I think it's not necessary to teach me how to test the TS, I've been studying this topic for a long time, and to explain for the 101st time that what you call MM, this is your generalization, and you do not even know that closing the profit in time is the same difficult task as not letting the losses to run out - I just can't tell it.
The highlighted shows a contradiction, which creates the problem.
It is advisable either to study the money management topic yourself or listen to your advices.
How to close the profit in time - only the analysis can explain
However, the MO turns off the market analyzing thoughts to the fullest extent and it also creates the basis of non-beliefs in the analysisthe highlighted shows a contradiction, which creates the problem
there is no problem, the only problem is your outlook, you just have not researched it and do not know where to look
It is advisable to either study the topic of money management yourself or listen to advice.
research
What advice? .... Shall we start a discussion about the state? - If not, there is no point in discussing it.
I don't think you are an expert in technical analysis, I'm not your client.
Why technical analysis? Is it just a word salad? - Why can't we find it by experience? - Where is the proof?
however MO turns off the market analyzing thoughts to the fullest extent and it also creates the basis for non-beliefs in the analysis
Or the state or again about the outlook - I don't even know, what are we discussing? .... Ah, yes I remember, "heavier-than-air vehicles can't fly" - so someone famous justified their vision of technical progress, and you're the same? Are you an expert in ME? - If you are a specialist in technical analysis, alas, I am not your client.