Yesterday our lead software engineer, Ilya Niklyaev, shared his impressions of his participation in the Kaggle machine learning competition at Singularis Lab Habr.com blog.
Always wanted to get acquainted with Machine Learning, but did not know how to start?
In his publication, Ilya talks about how he “right off the bat” - being mathematically savvy, but having no idea about Machine Learning, he took part in a serious ML competition from Kaggle!
The goal of the competition is to create a model that would be able to successfully predict the stock price of various companies based on the state of the asset and news related to the asset.
The article discusses all the steps that every participant have to follow before the finished model is ready:
-
How to come up with a way to locally validate the model;
-
How to choose the “correct” neural network architecture;
-
What additional steps need to be taken to improve the final result.
Read more by the link!