Created
June 17, 2020 00:07
-
-
Save dzakyputra/9c1269dde079a55eb9b6d630ba2a0269 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Profile preprocessing | |
| # Convert the became_member_on into datetime type | |
| profile['became_member_on'] = pd.to_datetime(profile['became_member_on'], format='%Y%m%d') | |
| # Create a new column with the value of the difference days between the column became_member_on and the max days | |
| profile['difference_days'] = (profile['became_member_on'].max() - profile['became_member_on']).dt.days | |
| ## Find the median and mode | |
| # Find median of age | |
| median_age_per_day = profile.groupby('became_member_on', as_index=False)['age'].median() | |
| # Find median of income | |
| median_income_per_day = profile.groupby('became_member_on', as_index=False)['income'].median() | |
| # Find mode of gender | |
| mode_gender_per_day = profile.groupby('became_member_on')['gender'].agg(lambda x: pd.Series.mode(x)) | |
| mode_gender_per_day_value = [i if isinstance(i, str) else 'M' for i in mode_gender_per_day] | |
| ## Fill the value based on the condition | |
| # Convert age 118 to the median of that day | |
| age_reference = dict(zip(median_age_per_day['became_member_on'], median_age_per_day['age'])) | |
| profile['age'] = profile['age'].replace({118: None, 101: None}).fillna(profile['became_member_on'].map(age_reference)) | |
| profile.loc[profile['age'] > 100, 'age'] = profile['age'].median() | |
| # Fill the null values in gender column with the mode | |
| gender_reference = dict(zip(mode_gender_per_day.index,mode_gender_per_day_value)) | |
| profile['gender'] = profile['gender'].fillna(profile['became_member_on'].map(gender_reference)) | |
| # Fill the null values in income column with the median | |
| income_reference = dict(zip(median_income_per_day['became_member_on'], median_income_per_day['income'])) | |
| profile['income'] = profile['income'].fillna(profile['became_member_on'].map(income_reference)) | |
| profile['income'].fillna(profile['income'].median(), inplace=True) | |
| # Round down the age column | |
| profile['age'] = profile['age'].astype(int) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment