search

mlp2018 / fraud_detection

Kaggle competition: TalkingData AdTracking Fraud Detection Challenge
Apache License 2.0
1 stars 0 forks source link

Import predictor variables from preprocessing.py #24

Closed johannadevos closed 6 years ago

johannadevos commented 6 years ago

Currently, lightgbm_main.py specifies the predictors that should be used:

# Columns our predictions are based on
    predictors = ['app', 'device', 'os', 'channel', 'hour', 'hour_sq', 'count_ip_day_freq_h', 'count_ip_day_hour', 'count_ip_hour_os', 'count_ip_hh_app', 'count_ip_hour_device']
    categorical = ['app', 'device', 'os', 'channel', 'hour', 'hour_sq', 'count_ip_day_freq_h', 'count_ip_day_hour', 'count_ip_hour_os', 'count_ip_hh_app', 'count_ip_hour_device']

However, this means that if André creates a new variable in preprocessing.py and wants it to be used as a predictor, lightgbm_main.py needs to be updated every time as well. Therefore, it might be a better idea if lightgbm_main.py would import the to-be-used predictors as a list from preprocessing.py.