Machine learning in trading: theory, models, practice and algo-trading - page 3418
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
No, 2 classes, but on different datasets with different objects. Merge them all into one.
First teach class 1 to Bai, class 2 to Sel, and then also class 1 to cats, class 2 to dogs? Seems like adding noise. We're almost 50/50 with no noise.....
First teach Bai to class 1, then Sel to class 2
I have seen several TSs that are set up separately by direction. Some of them trade only one side at all. I have no idea whether these directions differ from each other.
I have seen several TSs that are set up separately by directions. Some of them trade only one side at all. I have no idea whether these directions differ from each other.
Take a trend strategy, in one DC we run only Buy, in another only Sell.
In total, of course, there will be a minus, but it is likely that one of them will be a plus and the subscribers of signals will cover the loss :-) One of the signals is profitable, it trades in the existing trend :-)
and so - for forex instruments the Long Short difference is pejoratively small, only with stock exchange instruments it is present.
I have seen several TSs that are set up separately by directions. Some of them trade only one side at all. I have no idea whether these directions differ from each other.
In my opinion, this is the best, i.e. - memorising only one task with one model.
we take a trending strategy, in one DC we run only Buy, in the other only Sell.
In total, of course, there will be a minus, but with a probability in one of them plus and subscribers of signals will cover the loss :-)) One of the signals turned out to be profitable, trading in the existing trend :-)
and so - for forex instruments the difference Long Short is pejoratively small, only with stock exchange instruments it is present.
bullshit
1) what subscriber will subscribe to buy&hold with wild drawdowns and staggering around zero?
2) and if at the moment of opening two TS on the market will be a holobal flat, neither strategy will be in the plus....
Well, it's obvious, you need to think a little bit.
I was wondering what I train models on (mac M1).
and this is the whole chip :)
For 3 years the battery still lives as long (almost 24 hours in the mode of watching youtube). But according to tests has already died to 85%., began to work badly a couple of keys, which I successfully disassembled, put back together and nothing has changed. Well, this is generally the problem of all laptops.
The performance of the next generations (M2 and M3) didn't increase much (15-30%).
*In 3 years I haven't had a single glitch or freeze. Compare it with your vinda.
In my opinion this is the best, i.e. - memorising one model for only one task.
Intuitively for me it looks like fitting.
bullshit
1) what subscriber will subscribe to buy&hold with wild drawdowns and staggering around zero?
2) and if at the moment of opening two TS on the market will be a holobal flat, no strategy will be in the plus....
Well, it's obvious, you need to think a little bit.
There's a bot in the marketplace in the first place right now that just buys. But again from the category of: give me a good performance, otherwise nothing will work.
I had such a bot-conjuncturist, who only sold in the crisis, when everything was falling. He made a good profit.give me a good performance or nothing will work.