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
in a normal situation seed has almost no effect, it is the algorithm that matters. If you have to bother with seed, the data is already rubbish
checking on new data solves, if there are only 10 signs and not 1000, you can be sure about it to some extent.
I think the default depth is 6, it doesn't affect much either, except for critical values.
learning depth affects differently, depending on historical variability.
Yes, maybe on 4 seed predictors it doesn't affect much. The rubbish is a misconception. It is seed that essentially determines how many predictors will be used in the model.
All parameters are affected. I just wanted to say that you probably have an order of magnitude more combinations than examples. With 4 predictors I can see the sense in a model of 1-3 CB trees, with a learning rate of 0.3-0.5, otherwise it's already fitting.
You can just try to feed different samples to continue learning on new data. Even CatBoost seems to be able to do this. It also knows how to merge models, but I haven't looked into it.
It's gradient bousting...
which means that it learns from the errors of the previous one
and we only need to train on one model, and several times.
The only difference between the models is that the samples are shifted in time.
it's gradient bousting.....
that is, learning from the mistakes of the previous one
and we only need to train on one model, and several times.
The only difference between the models is that the samples are shifted in time.
My brain can't process what you've written.
Yeah, maybe on 4 predictors seed doesn't have much effect. The rubbish is a misconception. In fact, seed determines how many predictors will be used in the model.
All parameters are affected. I just wanted to say that you probably have an order of magnitude more combinations than examples. With 4 predictors I can see the sense in a model of 1-3 CB trees, with a learning rate of 0.3-0.5, otherwise it's already fitting.
The seed does not affect anywhere there is a normal optimum
+- short-circuit, it doesn't matter.
you can tweak it a little, but it's no longer crucial
seed has no effect anywhere there is a normal optimum
+- short-circuit, it does not play a role
You can tweak it a little, but it's not crucial.
Where is it?
And where is it?
where variations on the seed theme don't affect the result much, I guess )
where variations on the seed theme don't affect the result much, apparently )
Clearly not in our case...
Obviously not in our case.
Well, there's something to look forward to.
There is. But this is talking about the ideal world, sometimes it is better to adapt to the existing one.
It's a matter of whether it's better to poke around randomly or stick to a priori reliable information
Other than the start and end times (sessions, calendar) nothing comes to mind. What do you mean?