Discussing the article: "Data Science and ML (Part 33): Pandas Dataframe in MQL5, Data Collection for ML Usage made easier"

[MetaQuotes]

Check out the new article: Data Science and ML (Part 33): Pandas Dataframe in MQL5, Data Collection for ML Usage made easier.

When working with machine learning models, it’s essential to ensure consistency in the data used for training, validation, and testing. In this article, we will create our own version of the Pandas library in MQL5 to ensure a unified approach for handling machine learning data, for ensuring the same data is applied inside and outside MQL5, where most of the training occurs.

When it comes to working with machine learning models, it is essential that we have the same data structure if not the same values for all the environments; Training, validation, and testing. With Open Neural Network Exchange (ONNX) models being supported in MQL5 and MetaTrader 5, nowadays we have an opportunity to import models trained outside into the MQL5 language and use them for trading purposes.

Since most users use Python for training these Artificial Intelligence (AI) models which are then being deployed in MetaTrader 5 through MQL5 code, there could be a huge difference in how the data is organized, and often times even the values within the same data structure can be slightly different, this is due to the difference in the two technologies.

 

In this article, we are going to mimic the Pandas library available in Python language. It is one of the most popular libraries, useful particularly when it comes to working with vast amounts of data.

Since this library is used by data scientists to prepare and manipulate data used in training ML models, by harnessing its ability, we aim to have the same data playground in MQL5 as in Python.

Author: Omega J Msigwa