Open chews0n opened 3 years ago
Also good to think of whether or not we need to engineer a new feature and since our output is time dependant, potentially some previous time features that can be grabbed from the data already
heavily related to issue #15 as it is dependant or done at the same time
The following features have been assigned to each:
Hopefully with the code there already you're able to extract the data and it's somewhat clean. Please refer to the link below for a little bit of information re: feature engineering:
https://towardsdatascience.com/feature-engineering-for-machine-learning-3a5e293a5114