November 2019
Intermediate to advanced
346 pages
9h 36m
English
Having engineered promising new features, our next steps are to train-test split, process the data into a convenient time series form, and then classify. Our training and testing sets will be two temporal halves of the dataset. This way, we can easily ensure that the shape of the input for training is the same as the shape of the input for testing, without cheating in our evaluation.