Discussion of article "CatBoost machine learning algorithm from Yandex with no Python or R knowledge required"

MetaQuotes
Moderator
234929
MetaQuotes  

New article CatBoost machine learning algorithm from Yandex with no Python or R knowledge required has been published:

The article provides the code and the description of the main stages of the machine learning process using a specific example. To obtain the model, you do not need Python or R knowledge. Furthermore, basic MQL5 knowledge is enough — this is exactly my level. Therefore, I hope that the article will serve as a good tutorial for a broad audience, assisting those interested in evaluating machine learning capabilities and in implementing them in their programs.

The results are not very impressive, but it can be noted that the main trading rule "avoid money loss" is observed. Even if we choose another model from the CB_Svod.csv file, the effect would still be positive, because the financial result of the most unsuccessful model that we got is -25 points, and the average financial result of all models is 3889.9 points.


Fig. 9 Financial result of trained models for the period 01.08.2019 - 31.10.2020"


Author: Aleksey Vyazmikin

MyNameWasTaken
29
MyNameWasTaken  
I do not care if you say it is not impressive results.. what is impressive is your efforts and new thought you put into this article.. WELL DONE!!
Aleksey Vyazmikin
17732
Aleksey Vyazmikin  
MyNameWasTaken:
I do not care if you say it is not impressive results.. what is impressive is your efforts and new thought you put into this article.. WELL DONE!!

Thank you, I'm glad that this aroused interest.

PLAMEN VASILEV IVANOV
516
PLAMEN VASILEV IVANOV  

Hi Alex,

Good work!!! I plan to experiment based on your code. One question, do you normalize the predictors in your code somewhere (i had a quick look inside, but didn't notice), or it is done inside CatBoost? Some say normalization is a must for ML.

best regards

Aleksey Vyazmikin
17732
Aleksey Vyazmikin  
PLAMEN VASILEV IVANOV:

Hi Alex,

Good work!!! I plan to experiment based on your code. One question, do you normalize the predictors in your code somewhere (i had a quick look inside, but didn't notice), or it is done inside CatBoost? Some say normalization is a must for ML.

best regards

Hello. Normalization is required for neural networks. Tree-based models do not require normalization, since the partitioning of predictors occurs independently in the range from minus infinity to plus infinity, conditionally.

konorti
18
konorti  

Hi,

Thanks for the interesting work and article. I tried to repeat your work but when i start CB_Calc_Svod script it stops with memory leakage. model_mqh folder is not created. CB_svod.csv is generated but strangely my lines of the test and exam balans are filled with 0. Do you know how i can solve this issue? Thanks


Aleksey Vyazmikin
17732
Aleksey Vyazmikin  
konorti:

Hi,

Thanks for the interesting work and article. I tried to repeat your work but when i start CB_Calc_Svod script it stops with memory leakage. model_mqh folder is not created. CB_svod.csv is generated but strangely my lines of the test and exam balans are filled with 0. Do you know how i can solve this issue? Thanks


Hello. Did you run all the files after training?

  • _02_Rezultat_Exam.bat - settings for recording classification by the examination sample models
  • _02_Rezultat_test.bat - settings for recording classification by the testing sample models
  • _02_Rezultat_Train.bat — settings for recording classification by the learning sample models
  • _03_Metrik_Exam.bat — settings for recording the metrics of each tree of the examination sample models
  • _03_Metrik_Test.bat — settings for recording the metrics of each tree of the testing sample models
  • _03_Metrik_Train.bat — settings for recording the metrics of each tree of the training sample models
  • _04_Analiz_Exam.bat — settings for recording the assessment of predictor importance for the examination sample models
  • _04_Analiz_Test.bat — settings for recording the assessment of predictor importance for the testing sample models
  • _04_Analiz_Train.bat — settings for recording the assessment of predictor importance for the training sample models
konorti
18
konorti  

Aleksey Vyazmikin:

Hello. Did you run all the files after training?

  • _02_Rezultat_Exam.bat - settings for recording classification by the examination sample models
  • _02_Rezultat_test.bat - settings for recording classification by the testing sample models
  • _02_Rezultat_Train.bat — settings for recording classification by the learning sample models
  • _03_Metrik_Exam.bat — settings for recording the metrics of each tree of the examination sample models
  • _03_Metrik_Test.bat — settings for recording the metrics of each tree of the testing sample models
  • _03_Metrik_Train.bat — settings for recording the metrics of each tree of the training sample models
  • _04_Analiz_Exam.bat — settings for recording the assessment of predictor importance for the examination sample models
  • _04_Analiz_Test.bat — settings for recording the assessment of predictor importance for the testing sample models
  • _04_Analiz_Train.bat — settings for recording the assessment of predictor importance for the training sample models
Yes, I even waited after each to finish (first I started at the same time, as it was written in the article) 
Aleksey Vyazmikin
17732
Aleksey Vyazmikin  
konorti:
Yes, I even waited after each to finish (first I started at the same time, as it was written in the article) 

Show a screenshot of the directory with the model - you need to make sure that all the necessary files are there.


konorti
18
konorti  
Aleksey Vyazmikin:

Show a screenshot of the directory with the model - you need to make sure that all the necessary files are there.


here it is
Aleksey Vyazmikin
17732
Aleksey Vyazmikin  
konorti:
here it is

Please upload the log after the CB_Calc_Svod script is running.