Machine learning in trading: theory, models, practice and algo-trading - page 2971

New comment
 
On staites learning is not an option....
No robe.)
 
mytarmailS #:
On staites learning is not an option....
No robe.)

again midijorney contradicts you. It takes samples from real databases (from stats, from the past) and trains networks on them. The result is quite beautiful.

For NN to work, you need two things: 1) we need a reliable past, 2) we need a bunch of samples (either the best or the worst, or all of them, but with scores).

here we immediately get into the contradiction about "price includes everything"...and it doesn't :-) in point (2) we used information that is not available in (1).

 
Maxim Kuznetsov #:

again, midijorney contradicts you. There are samples taken from real bases (from stats, from the past), networks are trained on them

Well then, the grail has been learnt, what's keeping you from shoveling money)?
 
mytarmailS #:
Well then, the grail has been learnt, what's keeping you from raking in the dough?)

How about using your brain? :-))

Well, it's funny, 20 pages on "how to use the BLEEP".

 
Maxim Kuznetsov #:

maybe use your brain? :-)

Well, that's funny, 20 pages of "how to use the BLEEP".

Use it.
 
mytarmailS #:
Well then, the grail has been learnt, what's keeping you from shoveling money?)

What keeps us from shovelling money in this context is:

1) the lack of real samples. (samples, stats) in the available foreseeable space. Do you have friends in DC who will provide an "impersonal" client's state for 3-5-10 years ? (10 is fantastic, at least 3).

2) already mentioned, long real stats use information that is not available for repetition.

 
mytarmailS #:
For all my knowledge, I don't know how to algorithmise it, just have an understanding of the situation and that's it....
IO is like an indicator, and 99% of the intelligent robot is behind me...

1) It's either that, and it works.

2) Or it's all on automatic and it never works.


I'm still on point 1) but I'm dreaming of point 2)

not in terms of R attacks, but just that the inputs look like aliens.

in which there is almost nothing market (though what is market?) - just arithmetic.

Your ML entered from an overbuy/oversell at a strictly signified moment. Unity of place+time (which the branch is screwed on)

 
Maxim Dmitrievsky #:


That's why I chose it from the beginning and didn't miss out. Although I'm not a programmer and still in essence.

And who am I then))))

 
Maxim Dmitrievsky #:
The opposite situation. No one uses R from large implementers. And people are not the last in IT. Although python is not in great demand there, usually scala.

What ignorance? I originally downloaded R and python. I poked around in both and understood everything at once, without anyone even prompting me. On the contrary, an unbiased opinion. From R-studio in general left an unpleasant residue.

I used becky mailer until the last time, and not many people knew about it at all, emeditor instead of notepad is also not the most frequent choice, but these are tools, and language is a tool, what's the difference, what to make a craft with? The cuts are the main thing)))) Looks like my shovel is yellow, and it's better than your green one))))))

Although maybe not quite right, the size and configuration of shovels are different)))))

 
Valeriy Yastremskiy #:

what difference does it make to make a craft? The cuts are the main thing))))

Each job has its own tool, that's the difference...

because the cuts are the most important thing!


If you have to dig.

you can dig with a shovel, you can dig with a backhoe, and some people dig with a spoon.

Of course the excavator is the coolest.


If you need to analyse the soil chemically, why do you need a backhoe?

1...296429652966296729682969297029712972297329742975297629772978...3574
New comment