Machine learning in trading: theory, models, practice and algo-trading - page 3349

New comment
 
СанСаныч Фоменко #:

Your article doesn't have a graph for the "forward" mode of the tester, which is the real way to judge the model.

By the way, you use mashki, no matter what the difference with the price, and you should be careful with them, because under certain conditions of testing models by your own testers, as it is not funny and contradicts the whole TA, mashki look ahead. Using the "forward" mode, if there is looking ahead, you will get a big discrepancy in the results between the forward and the main plot.

You're hopeless.

I thought I asked you not to react to my posts then and not to take up my time, so they are not for you.
 
In the article, forward and backward are highlighted with vertical lines.

The article proposes to test the model in the MT5 tester, which does not have peeking as well as its own.

Try just training the model on the data. You will not get such a forward, i.e. the difference is already present. It is enough to check it yourself.

If you cannot understand the meaning with your head, you can at least see the difference. Including checking on your "super signs" that are "related" to the target.
 
Maxim Dmitrievsky with vertical lines.

The article proposes to test the model in the MT5 tester, which does not have peeking as well as its own.

Try just training the model on the data. You will not get such a forward, i.e. the difference is already present. It is enough to check it yourself.

If you cannot understand the meaning with your head, you can at least see the difference. Including checking on your "super features", which are "related" to the target one.

So many words... and all you have to do is to remove the bird in the tester.....

 
СанСаныч Фоменко #:

So many words. It's a simple matter of unchecking the bird in the tester...

So you don't perceive information at the level of words. Nor at the level of code. Not even at the level of formulas. I don't understand why you need to demonstrate this on a permanent basis :)

It's a question of cognitive ability. Why broadcast it, it looks cheesy, that is, it's like passing by.
 

Oh, the MO chip

Predicting Stock Price using Azure Automated Machine Learning (AutoML) in few clicks - TheWindowsUpdate.com

Predicting Stock Price using Azure Automated Machine Learning (AutoML) in few clicks
Predicting Stock Price using Azure Automated Machine Learning (AutoML) in few clicks
  • 2022.09.20
  • Syndicated News
  • thewindowsupdate.com
In a typical machine learning process, we first go through data featurization activities, including missing and imbalanced data handling exercises. We then select an appropriate algorithm to handle the business problems we are trying to solve through the machine learning model, followed by a hyper parameter tuning exercise to choose an optimal...
 
How much does it cost for 1,000 requests? You'd be broke on a tester.
 
Renat Akhtyamov #:

Oh, a chip on the MoD

Predicting Stock Price using Azure Automated Machine Learning (AutoML) in few clicks - TheWindowsUpdate.com

NOT more than a commercial offering in the MO field. There are a couple of hundred or more IO models that, when applied within the functionality outlined, do not allow trading for profit.

If we talk about the proposed functionality, we can compare it to the caret package, whose creators realised that the problem of MO is NOT in the models, but in the preprocessing and subsequent evaluation of the models.

 
MOs don't even know the difference between auto ml and full cyclic :)
If you make a model, your clients will start harassing you, and you pay for requests 😀😀 better your own server.
1...334233433344334533463347334833493350335133523353335433553356...3595
New comment