Skip to Content

Machine Learning

Last modified by Nikita Kapchenko on 2019/12/14 13:27
  1. Before training complex models, it is recommended to manipulate the data a bit, make some plots, and check simple assumptions. 
  2. A good starting point is to build 0, 1 or 2 deep decision trees and check their accuracy
  3. If it turns out that with enormous effort, we increase the share of correct answers by 0.5%, possibly we are doing something wrong, and it suffices to have a simple model with two conditions