Machine learning in trading: theory, models, practice and algo-trading - page 1115

 
Mihail Marchukajtes:

All true, but since that time I made a number of key changes that increased the amount of meaningful data. Especially I dumped a full set, and now after preprocessing the number of inputs decreases and only significant ones remain. You can try again to evaluate.... if you are interested!!!

Give me the file.

 
Mihail Marchukajtes:

Okay. I'll tell you right away that there are not many. I can make a giant unloading, but to do MT it will be long and I'm afraid will hang, so try to train it on what you have. I'll do the preprocessing myself and post the already preprocessed dataset. ok?

Don't forget datetime, so that I can break down the data correctly.

 
SanSanych Fomenko:

Give me the file.

I hope the result will not make you wait long, especially since there is not much data....

Files:
Buy_Test.txt  4 kb
Buy_Train.txt  3 kb
 
itslek:

Don't forget datetime, so that I can break down the data correctly.

Damn well, then for you separately will throw the full file without the breakdown and preprocessing. There are about 150 lines, for testing do at least 30-50 lines. I've never been able to get a model to work longer than that period... there will be the date and time...

I should just say right away that there is no concept of time for classification. At me the data before formation of sample are mixed by random. It's a well-known fact that way, if anything....

Files:
Full_Data.txt  225 kb
 
Brothers, just please post the result in the form of charts, so you can clearly assess, rather than dry numbers. OKYSHKI????
 
Vizard_:

itlek ran for Validol))))

Well, what about you???? spin them at your place, whatever way you go, but your opinion is also interesting.....

 
Vizard_:

itslek get Validol)))

He's too young for that.... green... Let's forgive him this omission :%-)

 
Mihail Marchukajtes:

Just to say right away that there is no concept of time for classification.

How is that?

That they are mixed, it is clear, I do not need to explain. Only about the time.

 

Wow, how the branch is empty.... It's obvious that people are at work :-)

And Mizgan posted kits.

♪ And the craft is turning ♪

♪ From the data, the nets swell the eyes ♪

And my head started spinning

♪ 'Cause the nets are looking for dogmas ♪

And each one found hers :-)

 
Yuriy Asaulenko:

What do you mean?

I don't need to explain that they mix. Just about the time.

What do you need time for, if the data are mixed before training, unless you want to isolate the OOS section... Don't worry, it's better to show results...