Machine learning in trading: theory, models, practice and algo-trading - page 1896
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
Depending on the hour (and obviously the instrument) there may be different cuts
Here, for example, the Eurobucks at the 9th hour (according to terminal time). Return cluster.
Buy and sell clusters can also be set to be positive on the history
in short, the question of overkill and optimization still remainsI see a picture from 5 min. The real result may be an inverted picture. Only from the real market. Doesn't matter demo or real. I will respect.
And so(((((.
I'm crying.))
I see a picture from 5min. The real result may be an inverted picture. Only from the real market give. Doesn't matter demo or real. I will respect.
And so(((((.
I'm crying.)))
Go take a walk.
Go take a walk.
Got it. So it's not working. Gone until better times).
Got it. So it doesn't work. Gone until better times).
It is not customary here to test with the real market. It's high time you understood the rules of the forum.
And this is correct. Period.
I see a picture from 5min. The real result may be an inverted picture. Only from the real market give. Doesn't matter demo or real. I will respect.
And so(((((.
I cried.))
The tester is like a drug.
For some people it is.
Self-soothing on the tester. Yes.
Don't want to stir up the anthill. I have been following this thread for a long time. No progress so far. Perhaps more progress behind the clouds.
"The deep learning algorithm with a teacher achieves acceptable quality with about 5,000 marked examples per category."
For m1 it's retraining every day on stories a week, for m5 once a week on stories a month.
For the other methods, are there any numbers to compare?
For some people, yes.
Self-soothing on the tester. Yeah.
I don't want to stir up the anthill. I've been following the thread for a long time. No progress so far. Perhaps there is still progress in the clouds.
Depending on the hour (and obviously the instrument) there may be different cuts
Here, for example, the Eurobucks at the 9th hour (according to terminal time). Return cluster.
Buy and sell clusters can also be set up to be positive on the history
In short, the question of search and optimization is still the same
+ retraining. On new data it can work worse (as usual)
Like this: since 2015 clustering. This is the most frustrating part. I didn't think a pretty rough clustering into clusters would give a lot of overtraining.
I still do not understand about time, 1, 2, 9 hour is just terminal time?