Machine learning in trading: theory, models, practice and algo-trading - page 2020
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
If you know how to cook, you can prepare...
Logically
the series has to be consistent. That is, not just torn from the history signs / marks, but a consistent series
I recommend that you just study the subject and try it yourself.
In my library, the data is transmitted as a 3-dimensional tensor. Each sample is a matrix. The whole dataset is divided into batches, which is another dimension.I wonder how the developers of all these neural networks, scaffolding, and boosts make money. They put everything out there for free.
The same way companies have been doing for 20-30 years developing Linux, the OS itself is free, but its support or fine-tuning to corporate requirements cost money and not small, all this is called service).
I wonder how the developers of all these neural networks, scaffolding, and boosts make money. They put everything out there for free.
Forests were invented in the 80s-90s and the first implementation on Fortran77 in the mid-80s. You can already give people to play with)
I wonder...
It started well again this week, and today it's not so good...
It's like this every start of the week... guessing the direction and then being stupid at the reversals
This week it started well again, and today it's not so good.
And so every beginning of the week... guesses the direction, and then stalls at the reversals
Maybe I should divide the beginning and the end of the week and teach separately) In general, it makes sense, Mondays, Tuesdays..... to gather separately and teach and use the results accordingly. Or when teaching not only take into account the increments, but also the days of the week.
It is clear, maybe divide the beginning of the week and the end of the week and teach separately) In general, it makes sense, Mondays, Tuesdays..... gather separately and teach and use the results accordingly. Or when teaching not only the increments, but also the days of the week.
I'll see what will happen next...
This week it started well again, and today it's not so good.
It's like this every start of the week... guessing the direction, and then getting dumb at the reversals.
It's interesting to see the balance over several weeks with the time frame...
And what do you mean by pre-training? Do you pre-train every weekend or what do you mean?
It's interesting to look at the balance over a few weeks with the timing...
And what do you mean by pre-training? Do you pre-train every weekend or what do you mean?
I run it in a tester like, and the same model is put into a trade, not a new one. Yes, in pond. I run the tester first, then the trader
I showed you with yellow dashes how it looks like every week.
I run in the type tester, and the same model is put into trade, not a new one. Yes, in pond. I run the tester first, then the trader
I showed you with yellow dashes how it looks like every week.
Well, then take the first two days and do not fuck your brains out ))
or try to pre-train online