Discussion of article "Thomas DeMark's Sequential (TD SEQUENTIAL) using artificial intelligence" - page 8
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
It's a shame it's rare to find articles on topics like this
Thank you! However, the material feels a certain understatement, something is missing :-(((((( I hope the next publication will dot the I's :-)
Yes, there are people but write what, I for example zero in coding, as experiments please, but I can not catch the essence and understand at what stage you are and what to achieve, and in general can you look at what neuronka if not difficult can then figure out what to add or what else to come up with.
Naturally I understand that you have progress as I understand after optimisation runs on non-optimiz. places results are and as I understand positive. Thank you I will wait for suggestions or something else.
I guess over 50 is good. But the dataset is skewed. When it is output balanced (equal number of ones and zeros) then the tester results will match your checks.
I meant which instrument what signs and what targets.
I meant what instrument, what signs and what goals.
In general, I trade exclusively Pound Dollar and the article is written for it, as for neurons, the models are in the files attached to the article.
I confess that I pursued this article with a vested interest, namely to find a programmer and jointly test the theory on his AI with my data. But as it turned out, either the article was not liked, or as suggested by the curator of the article that few people can support the conversation. Therefore I urge readers to discuss what is described in the article, maybe something is not clear, maybe somewhere it is necessary to clarify in more detail????. Do not hesitate to ask, always happy to support the conversation. And remember that the stupidest question is the one that is not asked.
Rather, the point is that market programmers (they are also traders) have their own thoughts like dirt.))) You can't check everything.)
I have already taken something from your article, something from SanSanych's article, most of it from publicly available sources on Data Mining, and I am working on something not yet fully clear with an unclear result). And it does not resemble your or SanSanych's systems in any way.
And to chew things up - then you have to start from the basics, which is in no way the competence of the MQL site. There are plenty of popular articles in the Internet. And not only popular ones.
Thank you for the article.
Rather, the point is that market programmers (they are also traders) have their own thoughts like dirt.))) You can't check everything.)
I have already taken something from your article, something from SanSanych's article, most of it from publicly available sources on Data Mining, and I am working on something not yet fully clear with unclear results). And it does not resemble your or SanSanych's systems in any way.
And to chew things up - then you have to start from the basics, which is in no way the competence of the MQL site. There are plenty of popular articles in the Internet. And not only popular ones.
Thank you for the article.
I am glad to hear it. Thank you! It's just that, as I've already said, I'm a specialist in AI training, not its development. So when you have any questions about data collection, target selection, what you can use and what you can't use in such and such a form. Contact me, always glad to help!!!!!
That's good to hear. Thank you. It's just that, as I said, I'm a specialist in AI training, not AI development. So when you have any questions about data collection, target selection, what can be used and what can't be used in such and such a form. Contact me, always glad to help!!!!!
Michael as I understand you need to code your development, as you need to conduct training and trading, if there is a problem with the programmer I can help you to create such an advisor, I will be glad to work with you for a good cause, leave skype there it will be easier to talk.
Michael as I understand you need to code your development, as you need to conduct training and trading, if there is a problem with the programmer I can help you to create such an advisor, I will be glad to work with you for a good cause, leave skype there it will be easier to talk.
To be honest, I really need help. I need to modify this indicator to make it work in the tester. In fact, the indicator takes data from the file (it was already in it), I just added a buffer to which I write the value of the difference between the current and previous value. When saving, everything works correctly, the data is unloaded for the required days, but it does not want to work in real time. Somewhere an error or I did something wrong. In general, I need to work in the tester. Do you think it's possible?