We will look at the development cycle of a trading robot: data collection, processing, sample expansion, feature engineering, model selection and training, creating a trading system via Python, and monitoring trades. Working in Python has its own advantages: speed in the field of machine learning, as well as the ability to select and generate features. Exporting models to ONNX requires exactly the same feature generation logic as in Python, which is not easy. That is why I have selected ...
In the previous article entitled "Practical application of neural networks in trading. It's time to practice", we considered the practical application of a neural network module implemented using Matlab neural networks. However, that article did not cover questions related to the preparation of input data and network training related operations. In this article, we will consider these questions using examples and will implement further code using ...
Gradient boosting is a powerful machine learning algorithm. The method produces an ensemble of weak models (for example, decision trees), in which (in contrast to bagging) models are built sequentially, rather than independently (in parallel). This means that the next tree learns from the mistakes of the previous one, then this process is repeated, increasing the number of weak models. This builds a strong model which can generalize using heterogeneous ...