is_mall Feature - Githubissues

SimenQ / machine_learning

Machine learning project

0 stars 0 forks source link

is_mall Feature #9

Open nicolaswon opened 2 years ago

nicolaswon commented 2 years ago

Description Boolean value of whether a store is a part of a mall

Details Might be helpful to do some data cleaning. I.e. if coordinates of stores are in the same location of a mall, but the mall feature is missing -> impude.

Datasets

Stores_train (primary)
Stores_extra (secondary)

ArashAminpour commented 2 years ago

Name feature: Is_mall

Need to check if the stores in stores_train have missing value in column MallName since they are not in a Mall or that they have missing value but is actually a part of a Mall.

- Check the stores_train dataset for GK id with NaN in mall-coloum
- Check for corresponding GK id in stores_extra
- If mallName exist in stores_extra at the corresponding Long Lat -> use the fillna() function to fill the missing value in stores_train
- Iterate through stores_train column MallName
- If value = NaN => Is_mall = False
- Else -> Is_mall = True