That's a good question. `df_feats_validation` contains exactly the same features as `df_feats`, only for points in validation time interval. Yes I probably could put the whole code together to give better understanding. There won't be much new code though, as most of it is already included in the article, just fragmented.
Software Architect, Data Scientist, Kaggle Grandmaster (https://www.kaggle.com/alijs1). Sometimes I write about Python, Data Science and Machine Learning…