Machine learning in trading: theory, models, practice and algo-trading - page 2226

 
Aleksey Nikolayev:
If I see hints on this forum about possessing the Great Secrets of the Trade, I always recall a funny example described by Freud (I think it was in his book "Libido"). He described his patient, a teenager, who claimed to have made a terrible discovery that would turn the world upside down. In the end, this "great discovery" turned out to be masturbation)

the thing is, you don't have to remember it, because mo's and other crap is just masturbation

You should only look and count your own money, and Max showed the picture above, by which even a blind man can see that Equity already reflects price movements

If you do not know anything else, why do you need to analyze it?

If you have learned to manage, count and trade using only your capital, you may not call what is going on in any other way.
 
Aleksey Nikolayev:
If I see hints on this forum about possessing the Great Secrets of the Trade, I always recall a funny example described by Freud (I think it was in his book "Libido"). He described his patient, a teenager, who claimed to have made a terrible discovery that would turn the world upside down. In the end, this "great discovery" turned out to be masturbation)
:)))))
 
Maxim Dmitrievsky:
🤣🤣🤣

I do not see these squares (that's how I see the comment).

 
Aleksey Vyazmikin:

I do not see these squares (so I see the comment).

Reworked ) it smiley faces from an iPhone, maybe only I see them
 

https://youtu.be/qlFTxADGixA?t=3407

it turns out that Yandex, with its matrixnet in the form of catbust, has already switched to transformers

it's time to transfer the bots to them )

YaC 2020: как мы делаем Яндекс
YaC 2020: как мы делаем Яндекс
  • 2020.11.25
  • www.youtube.com
Каждый год Яндекс устраивает большую технологическую конференцию, но в этом году мы поменяли формат и сняли фильм. Получилась обстоятельная беседа управляющ...
 
Maxim Dmitrievsky:

https://youtu.be/qlFTxADGixA?t=3407

it turns out that Yandex, with its matrixnet in the form of catbust, has already switched to transformers

It's time to transfer the bots to them )

Are these transformers an open project?

 
Aleksey Vyazmikin:

So these Transformers are an open project?

https://habr.com/ru/company/yandex/blog/529658/

they are open architecture for working with sequences, examples are in all packages

I've written about them 500 times already, I can't find time to test them.

You know, Yandex has been working on it for years. How much time it takes to add it for trading. If there were strong machinists here who know pythorch, they could organize a research co-op.

Трансформеры в Поиске: как Яндекс применил тяжёлые нейросети для поиска по смыслу
Трансформеры в Поиске: как Яндекс применил тяжёлые нейросети для поиска по смыслу
  • habr.com
Привет, Хабр. Меня зовут Саша Готманов, я руковожу группой нейросетевых технологий в поиске Яндекса. На YaC 2020 мы впервые рассказали о внедрении трансформера — новой нейросетевой архитектуры для ранжирования веб-страниц. Это наиболее значимое событие в нашем поиске за последние 10 лет.  Сегодня я расскажу читателям Хабра, в чём заключается...
 
Maxim Dmitrievsky:

https://habr.com/ru/company/yandex/blog/529658/

these are open sequencing architectures, examples are in all packages

I've written about them 500 times already, I can't find time to test them.

The theory is clear - the question is in realization, and it's closed - only pictures, and the sharpening is for text.

 
Aleksey Vyazmikin:

Yes theory is clear - the question is in the implementation, and it is closed - only pictures, and sharpening there under the text.

ugh shit... google helps... or yandex

 
Maxim Dmitrievsky:

Oh shit... google help... or Yandex

I mean the specific implementation from Yandex. In python it is clear that everything is not complicated from publicly available options :)