These columns indicate those that contain imputed data. For the most part, we have included the unimputed counterparts in our data and therefore do not need the imputed columns, so we remove them:
imp_cols = [
'AGEFL','BDATEFL','SEXFL','ETHNICFL','RACERFL'
]
X_train.drop(imp_cols, axis=1, inplace=True)
X_test.drop(imp_cols, axis=1, inplace=True)