Déploiement automatique depuis GitLab CI - 2026-02-20 09:00:53

2_Data_transformed/app_train_Enc_wo_Outliers_Feat_Eng_Join_Align_head.csv

@@ -0,0 +1,2 @@
+SK_ID_CURR,CODE_GENDER,FLAG_OWN_CAR,FLAG_OWN_REALTY,CNT_CHILDREN,AMT_INCOME_TOTAL,AMT_CREDIT,AMT_ANNUITY,AMT_GOODS_PRICE,REGION_POPULATION_RELATIVE,DAYS_BIRTH,DAYS_EMPLOYED,DAYS_REGISTRATION,DAYS_ID_PUBLISH,OWN_CAR_AGE,FLAG_MOBIL,FLAG_EMP_PHONE,FLAG_WORK_PHONE,FLAG_CONT_MOBILE,FLAG_PHONE,FLAG_EMAIL,CNT_FAM_MEMBERS,REGION_RATING_CLIENT,REGION_RATING_CLIENT_W_CITY,HOUR_APPR_PROCESS_START,REG_REGION_NOT_LIVE_REGION,REG_REGION_NOT_WORK_REGION,LIVE_REGION_NOT_WORK_REGION,REG_CITY_NOT_LIVE_CITY,REG_CITY_NOT_WORK_CITY,LIVE_CITY_NOT_WORK_CITY,EXT_SOURCE_1,EXT_SOURCE_2,EXT_SOURCE_3,APARTMENTS_AVG,BASEMENTAREA_AVG,YEARS_BEGINEXPLUATATION_AVG,YEARS_BUILD_AVG,COMMONAREA_AVG,ELEVATORS_AVG,ENTRANCES_AVG,FLOORSMAX_AVG,FLOORSMIN_AVG,LANDAREA_AVG,LIVINGAPARTMENTS_AVG,LIVINGAREA_AVG,NONLIVINGAPARTMENTS_AVG,NONLIVINGAREA_AVG,APARTMENTS_MODE,BASEMENTAREA_MODE,YEARS_BEGINEXPLUATATION_MODE,YEARS_BUILD_MODE,COMMONAREA_MODE,ELEVATORS_MODE,ENTRANCES_MODE,FLOORSMAX_MODE,FLOORSMIN_MODE,LANDAREA_MODE,LIVINGAPARTMENTS_MODE,LIVINGAREA_MODE,NONLIVINGAPARTMENTS_MODE,NONLIVINGAREA_MODE,APARTMENTS_MEDI,BASEMENTAREA_MEDI,YEARS_BEGINEXPLUATATION_MEDI,YEARS_BUILD_MEDI,COMMONAREA_MEDI,ELEVATORS_MEDI,ENTRANCES_MEDI,FLOORSMAX_MEDI,FLOORSMIN_MEDI,LANDAREA_MEDI,LIVINGAPARTMENTS_MEDI,LIVINGAREA_MEDI,NONLIVINGAPARTMENTS_MEDI,NONLIVINGAREA_MEDI,TOTALAREA_MODE,OBS_30_CNT_SOCIAL_CIRCLE,DEF_30_CNT_SOCIAL_CIRCLE,OBS_60_CNT_SOCIAL_CIRCLE,DEF_60_CNT_SOCIAL_CIRCLE,DAYS_LAST_PHONE_CHANGE,FLAG_DOCUMENT_2,FLAG_DOCUMENT_3,FLAG_DOCUMENT_4,FLAG_DOCUMENT_5,FLAG_DOCUMENT_6,FLAG_DOCUMENT_7,FLAG_DOCUMENT_8,FLAG_DOCUMENT_9,FLAG_DOCUMENT_10,FLAG_DOCUMENT_11,FLAG_DOCUMENT_12,FLAG_DOCUMENT_13,FLAG_DOCUMENT_14,FLAG_DOCUMENT_15,FLAG_DOCUMENT_16,FLAG_DOCUMENT_17,FLAG_DOCUMENT_18,FLAG_DOCUMENT_19,FLAG_DOCUMENT_20,FLAG_DOCUMENT_21,AMT_REQ_CREDIT_BUREAU_HOUR,AMT_REQ_CREDIT_BUREAU_DAY,AMT_REQ_CREDIT_BUREAU_WEEK,AMT_REQ_CREDIT_BUREAU_MON,AMT_REQ_CREDIT_BUREAU_QRT,AMT_REQ_CREDIT_BUREAU_YEAR,NAME_CONTRACT_TYPE_Cash loans,NAME_CONTRACT_TYPE_Revolving loans,NAME_TYPE_SUITE_Children,NAME_TYPE_SUITE_Family,NAME_TYPE_SUITE_Group of people,NAME_TYPE_SUITE_Other_A,NAME_TYPE_SUITE_Other_B,"NAME_TYPE_SUITE_Spouse, partner",NAME_TYPE_SUITE_Unaccompanied,NAME_INCOME_TYPE_Businessman,NAME_INCOME_TYPE_Commercial associate,NAME_INCOME_TYPE_Pensioner,NAME_INCOME_TYPE_State servant,NAME_INCOME_TYPE_Student,NAME_INCOME_TYPE_Unemployed,NAME_INCOME_TYPE_Working,NAME_EDUCATION_TYPE_Academic degree,NAME_EDUCATION_TYPE_Higher education,NAME_EDUCATION_TYPE_Incomplete higher,NAME_EDUCATION_TYPE_Lower secondary,NAME_EDUCATION_TYPE_Secondary / secondary special,NAME_FAMILY_STATUS_Civil marriage,NAME_FAMILY_STATUS_Married,NAME_FAMILY_STATUS_Separated,NAME_FAMILY_STATUS_Single / not married,NAME_FAMILY_STATUS_Widow,NAME_HOUSING_TYPE_Co-op apartment,NAME_HOUSING_TYPE_House / apartment,NAME_HOUSING_TYPE_Municipal apartment,NAME_HOUSING_TYPE_Office apartment,NAME_HOUSING_TYPE_Rented apartment,NAME_HOUSING_TYPE_With parents,OCCUPATION_TYPE_Accountants,OCCUPATION_TYPE_Cleaning staff,OCCUPATION_TYPE_Cooking staff,OCCUPATION_TYPE_Core staff,OCCUPATION_TYPE_Drivers,OCCUPATION_TYPE_HR staff,OCCUPATION_TYPE_High skill tech staff,OCCUPATION_TYPE_IT staff,OCCUPATION_TYPE_Laborers,OCCUPATION_TYPE_Low-skill Laborers,OCCUPATION_TYPE_Managers,OCCUPATION_TYPE_Medicine staff,OCCUPATION_TYPE_Private service staff,OCCUPATION_TYPE_Realty agents,OCCUPATION_TYPE_Sales staff,OCCUPATION_TYPE_Secretaries,OCCUPATION_TYPE_Security staff,OCCUPATION_TYPE_Waiters/barmen staff,WEEKDAY_APPR_PROCESS_START_FRIDAY,WEEKDAY_APPR_PROCESS_START_MONDAY,WEEKDAY_APPR_PROCESS_START_SATURDAY,WEEKDAY_APPR_PROCESS_START_SUNDAY,WEEKDAY_APPR_PROCESS_START_THURSDAY,WEEKDAY_APPR_PROCESS_START_TUESDAY,WEEKDAY_APPR_PROCESS_START_WEDNESDAY,ORGANIZATION_TYPE_Advertising,ORGANIZATION_TYPE_Agriculture,ORGANIZATION_TYPE_Bank,ORGANIZATION_TYPE_Business Entity Type 1,ORGANIZATION_TYPE_Business Entity Type 2,ORGANIZATION_TYPE_Business Entity Type 3,ORGANIZATION_TYPE_Cleaning,ORGANIZATION_TYPE_Construction,ORGANIZATION_TYPE_Culture,ORGANIZATION_TYPE_Electricity,ORGANIZATION_TYPE_Emergency,ORGANIZATION_TYPE_Government,ORGANIZATION_TYPE_Hotel,ORGANIZATION_TYPE_Housing,ORGANIZATION_TYPE_Industry: type 1,ORGANIZATION_TYPE_Industry: type 10,ORGANIZATION_TYPE_Industry: type 11,ORGANIZATION_TYPE_Industry: type 12,ORGANIZATION_TYPE_Industry: type 13,ORGANIZATION_TYPE_Industry: type 2,ORGANIZATION_TYPE_Industry: type 3,ORGANIZATION_TYPE_Industry: type 4,ORGANIZATION_TYPE_Industry: type 5,ORGANIZATION_TYPE_Industry: type 6,ORGANIZATION_TYPE_Industry: type 7,ORGANIZATION_TYPE_Industry: type 8,ORGANIZATION_TYPE_Industry: type 9,ORGANIZATION_TYPE_Insurance,ORGANIZATION_TYPE_Kindergarten,ORGANIZATION_TYPE_Legal Services,ORGANIZATION_TYPE_Medicine,ORGANIZATION_TYPE_Military,ORGANIZATION_TYPE_Mobile,ORGANIZATION_TYPE_Other,ORGANIZATION_TYPE_Police,ORGANIZATION_TYPE_Postal,ORGANIZATION_TYPE_Realtor,ORGANIZATION_TYPE_Religion,ORGANIZATION_TYPE_Restaurant,ORGANIZATION_TYPE_School,ORGANIZATION_TYPE_Security,ORGANIZATION_TYPE_Security Ministries,ORGANIZATION_TYPE_Self-employed,ORGANIZATION_TYPE_Services,ORGANIZATION_TYPE_Telecom,ORGANIZATION_TYPE_Trade: type 1,ORGANIZATION_TYPE_Trade: type 2,ORGANIZATION_TYPE_Trade: type 3,ORGANIZATION_TYPE_Trade: type 4,ORGANIZATION_TYPE_Trade: type 5,ORGANIZATION_TYPE_Trade: type 6,ORGANIZATION_TYPE_Trade: type 7,ORGANIZATION_TYPE_Transport: type 1,ORGANIZATION_TYPE_Transport: type 2,ORGANIZATION_TYPE_Transport: type 3,ORGANIZATION_TYPE_Transport: type 4,ORGANIZATION_TYPE_University,ORGANIZATION_TYPE_XNA,FONDKAPREMONT_MODE_not specified,FONDKAPREMONT_MODE_org spec account,FONDKAPREMONT_MODE_reg oper account,FONDKAPREMONT_MODE_reg oper spec account,HOUSETYPE_MODE_block of flats,HOUSETYPE_MODE_specific housing,HOUSETYPE_MODE_terraced house,WALLSMATERIAL_MODE_Block,WALLSMATERIAL_MODE_Mixed,WALLSMATERIAL_MODE_Monolithic,WALLSMATERIAL_MODE_Others,WALLSMATERIAL_MODE_Panel,"WALLSMATERIAL_MODE_Stone, brick",WALLSMATERIAL_MODE_Wooden,EMERGENCYSTATE_MODE_No,EMERGENCYSTATE_MODE_Yes,DAYS_EMPLOYED_PERC,INCOME_CREDIT_PERC,INCOME_PER_PERSON,ANNUITY_INCOME_PERC,PAYMENT_RATE,BURO_DAYS_CREDIT_MIN,BURO_DAYS_CREDIT_MAX,BURO_DAYS_CREDIT_MEAN,BURO_DAYS_CREDIT_VAR,BURO_DAYS_CREDIT_ENDDATE_MIN,BURO_DAYS_CREDIT_ENDDATE_MAX,BURO_DAYS_CREDIT_ENDDATE_MEAN,BURO_DAYS_CREDIT_UPDATE_MEAN,BURO_CREDIT_DAY_OVERDUE_MAX,BURO_CREDIT_DAY_OVERDUE_MEAN,BURO_AMT_CREDIT_MAX_OVERDUE_MEAN,BURO_AMT_CREDIT_SUM_MAX,BURO_AMT_CREDIT_SUM_MEAN,BURO_AMT_CREDIT_SUM_SUM,BURO_AMT_CREDIT_SUM_DEBT_MAX,BURO_AMT_CREDIT_SUM_DEBT_MEAN,BURO_AMT_CREDIT_SUM_DEBT_SUM,BURO_AMT_CREDIT_SUM_OVERDUE_MEAN,BURO_AMT_CREDIT_SUM_LIMIT_MEAN,BURO_AMT_CREDIT_SUM_LIMIT_SUM,BURO_AMT_ANNUITY_MAX,BURO_AMT_ANNUITY_MEAN,BURO_CNT_CREDIT_PROLONG_SUM,BURO_MONTHS_BALANCE_MIN_MIN,BURO_MONTHS_BALANCE_MAX_MAX,BURO_MONTHS_BALANCE_SIZE_MEAN,BURO_MONTHS_BALANCE_SIZE_SUM,BURO_CREDIT_ACTIVE_Active_MEAN,BURO_CREDIT_ACTIVE_Bad debt_MEAN,BURO_CREDIT_ACTIVE_Closed_MEAN,BURO_CREDIT_ACTIVE_Sold_MEAN,BURO_CREDIT_ACTIVE_nan_MEAN,BURO_CREDIT_CURRENCY_currency 1_MEAN,BURO_CREDIT_CURRENCY_currency 2_MEAN,BURO_CREDIT_CURRENCY_currency 3_MEAN,BURO_CREDIT_CURRENCY_currency 4_MEAN,BURO_CREDIT_CURRENCY_nan_MEAN,BURO_CREDIT_TYPE_Another type of loan_MEAN,BURO_CREDIT_TYPE_Car loan_MEAN,BURO_CREDIT_TYPE_Cash loan (non-earmarked)_MEAN,BURO_CREDIT_TYPE_Consumer credit_MEAN,BURO_CREDIT_TYPE_Credit card_MEAN,BURO_CREDIT_TYPE_Interbank credit_MEAN,BURO_CREDIT_TYPE_Loan for business development_MEAN,BURO_CREDIT_TYPE_Loan for purchase of shares (margin lending)_MEAN,BURO_CREDIT_TYPE_Loan for the purchase of equipment_MEAN,BURO_CREDIT_TYPE_Loan for working capital replenishment_MEAN,BURO_CREDIT_TYPE_Microloan_MEAN,BURO_CREDIT_TYPE_Mobile operator loan_MEAN,BURO_CREDIT_TYPE_Mortgage_MEAN,BURO_CREDIT_TYPE_Real estate loan_MEAN,BURO_CREDIT_TYPE_Unknown type of loan_MEAN,BURO_CREDIT_TYPE_nan_MEAN,BURO_STATUS_0_MEAN_MEAN,BURO_STATUS_1_MEAN_MEAN,BURO_STATUS_2_MEAN_MEAN,BURO_STATUS_3_MEAN_MEAN,BURO_STATUS_4_MEAN_MEAN,BURO_STATUS_5_MEAN_MEAN,BURO_STATUS_C_MEAN_MEAN,BURO_STATUS_X_MEAN_MEAN,BURO_STATUS_nan_MEAN_MEAN,ACTIVE_DAYS_CREDIT_MIN,ACTIVE_DAYS_CREDIT_MAX,ACTIVE_DAYS_CREDIT_MEAN,ACTIVE_DAYS_CREDIT_VAR,ACTIVE_DAYS_CREDIT_ENDDATE_MIN,ACTIVE_DAYS_CREDIT_ENDDATE_MAX,ACTIVE_DAYS_CREDIT_ENDDATE_MEAN,ACTIVE_DAYS_CREDIT_UPDATE_MEAN,ACTIVE_CREDIT_DAY_OVERDUE_MAX,ACTIVE_CREDIT_DAY_OVERDUE_MEAN,ACTIVE_AMT_CREDIT_MAX_OVERDUE_MEAN,ACTIVE_AMT_CREDIT_SUM_MAX,ACTIVE_AMT_CREDIT_SUM_MEAN,ACTIVE_AMT_CREDIT_SUM_SUM,ACTIVE_AMT_CREDIT_SUM_DEBT_MAX,ACTIVE_AMT_CREDIT_SUM_DEBT_MEAN,ACTIVE_AMT_CREDIT_SUM_DEBT_SUM,ACTIVE_AMT_CREDIT_SUM_OVERDUE_MEAN,ACTIVE_AMT_CREDIT_SUM_LIMIT_MEAN,ACTIVE_AMT_CREDIT_SUM_LIMIT_SUM,ACTIVE_AMT_ANNUITY_MAX,ACTIVE_AMT_ANNUITY_MEAN,ACTIVE_CNT_CREDIT_PROLONG_SUM,ACTIVE_MONTHS_BALANCE_MIN_MIN,ACTIVE_MONTHS_BALANCE_MAX_MAX,ACTIVE_MONTHS_BALANCE_SIZE_MEAN,ACTIVE_MONTHS_BALANCE_SIZE_SUM,CLOSED_DAYS_CREDIT_MIN,CLOSED_DAYS_CREDIT_MAX,CLOSED_DAYS_CREDIT_MEAN,CLOSED_DAYS_CREDIT_VAR,CLOSED_DAYS_CREDIT_ENDDATE_MIN,CLOSED_DAYS_CREDIT_ENDDATE_MAX,CLOSED_DAYS_CREDIT_ENDDATE_MEAN,CLOSED_DAYS_CREDIT_UPDATE_MEAN,CLOSED_CREDIT_DAY_OVERDUE_MAX,CLOSED_CREDIT_DAY_OVERDUE_MEAN,CLOSED_AMT_CREDIT_MAX_OVERDUE_MEAN,CLOSED_AMT_CREDIT_SUM_MAX,CLOSED_AMT_CREDIT_SUM_MEAN,CLOSED_AMT_CREDIT_SUM_SUM,CLOSED_AMT_CREDIT_SUM_DEBT_MAX,CLOSED_AMT_CREDIT_SUM_DEBT_MEAN,CLOSED_AMT_CREDIT_SUM_DEBT_SUM,CLOSED_AMT_CREDIT_SUM_OVERDUE_MEAN,CLOSED_AMT_CREDIT_SUM_LIMIT_MEAN,CLOSED_AMT_CREDIT_SUM_LIMIT_SUM,CLOSED_AMT_ANNUITY_MAX,CLOSED_AMT_ANNUITY_MEAN,CLOSED_CNT_CREDIT_PROLONG_SUM,CLOSED_MONTHS_BALANCE_MIN_MIN,CLOSED_MONTHS_BALANCE_MAX_MAX,CLOSED_MONTHS_BALANCE_SIZE_MEAN,CLOSED_MONTHS_BALANCE_SIZE_SUM,PREV_AMT_ANNUITY_MIN,PREV_AMT_ANNUITY_MAX,PREV_AMT_ANNUITY_MEAN,PREV_AMT_APPLICATION_MIN,PREV_AMT_APPLICATION_MAX,PREV_AMT_APPLICATION_MEAN,PREV_AMT_CREDIT_MIN,PREV_AMT_CREDIT_MAX,PREV_AMT_CREDIT_MEAN,PREV_APP_CREDIT_PERC_MIN,PREV_APP_CREDIT_PERC_MAX,PREV_APP_CREDIT_PERC_MEAN,PREV_APP_CREDIT_PERC_VAR,PREV_AMT_DOWN_PAYMENT_MIN,PREV_AMT_DOWN_PAYMENT_MAX,PREV_AMT_DOWN_PAYMENT_MEAN,PREV_AMT_GOODS_PRICE_MIN,PREV_AMT_GOODS_PRICE_MAX,PREV_AMT_GOODS_PRICE_MEAN,PREV_HOUR_APPR_PROCESS_START_MIN,PREV_HOUR_APPR_PROCESS_START_MAX,PREV_HOUR_APPR_PROCESS_START_MEAN,PREV_RATE_DOWN_PAYMENT_MIN,PREV_RATE_DOWN_PAYMENT_MAX,PREV_RATE_DOWN_PAYMENT_MEAN,PREV_DAYS_DECISION_MIN,PREV_DAYS_DECISION_MAX,PREV_DAYS_DECISION_MEAN,PREV_CNT_PAYMENT_MEAN,PREV_CNT_PAYMENT_SUM,PREV_NAME_CONTRACT_TYPE_Cash loans_MEAN,PREV_NAME_CONTRACT_TYPE_Consumer loans_MEAN,PREV_NAME_CONTRACT_TYPE_Revolving loans_MEAN,PREV_NAME_CONTRACT_TYPE_XNA_MEAN,PREV_NAME_CONTRACT_TYPE_nan_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_FRIDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_MONDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_SATURDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_SUNDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_THURSDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_TUESDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_WEDNESDAY_MEAN,PREV_WEEKDAY_APPR_PROCESS_START_nan_MEAN,PREV_FLAG_LAST_APPL_PER_CONTRACT_N_MEAN,PREV_FLAG_LAST_APPL_PER_CONTRACT_Y_MEAN,PREV_FLAG_LAST_APPL_PER_CONTRACT_nan_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Building a house or an annex_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Business development_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Buying a garage_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Buying a holiday home / land_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Buying a home_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Buying a new car_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Buying a used car_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Car repairs_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Education_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Everyday expenses_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Furniture_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Gasification / water supply_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Hobby_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Journey_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Medicine_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Money for a third person_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Other_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Payments on other loans_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Purchase of electronic equipment_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Refusal to name the goal_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Repairs_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Urgent needs_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_Wedding / gift / holiday_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_XAP_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_XNA_MEAN,PREV_NAME_CASH_LOAN_PURPOSE_nan_MEAN,PREV_NAME_CONTRACT_STATUS_Approved_MEAN,PREV_NAME_CONTRACT_STATUS_Canceled_MEAN,PREV_NAME_CONTRACT_STATUS_Refused_MEAN,PREV_NAME_CONTRACT_STATUS_Unused offer_MEAN,PREV_NAME_CONTRACT_STATUS_nan_MEAN,PREV_NAME_PAYMENT_TYPE_Cash through the bank_MEAN,PREV_NAME_PAYMENT_TYPE_Cashless from the account of the employer_MEAN,PREV_NAME_PAYMENT_TYPE_Non-cash from your account_MEAN,PREV_NAME_PAYMENT_TYPE_XNA_MEAN,PREV_NAME_PAYMENT_TYPE_nan_MEAN,PREV_CODE_REJECT_REASON_CLIENT_MEAN,PREV_CODE_REJECT_REASON_HC_MEAN,PREV_CODE_REJECT_REASON_LIMIT_MEAN,PREV_CODE_REJECT_REASON_SCO_MEAN,PREV_CODE_REJECT_REASON_SCOFR_MEAN,PREV_CODE_REJECT_REASON_SYSTEM_MEAN,PREV_CODE_REJECT_REASON_VERIF_MEAN,PREV_CODE_REJECT_REASON_XAP_MEAN,PREV_CODE_REJECT_REASON_XNA_MEAN,PREV_CODE_REJECT_REASON_nan_MEAN,PREV_NAME_TYPE_SUITE_Children_MEAN,PREV_NAME_TYPE_SUITE_Family_MEAN,PREV_NAME_TYPE_SUITE_Group of people_MEAN,PREV_NAME_TYPE_SUITE_Other_A_MEAN,PREV_NAME_TYPE_SUITE_Other_B_MEAN,"PREV_NAME_TYPE_SUITE_Spouse, partner_MEAN",PREV_NAME_TYPE_SUITE_Unaccompanied_MEAN,PREV_NAME_TYPE_SUITE_nan_MEAN,PREV_NAME_CLIENT_TYPE_New_MEAN,PREV_NAME_CLIENT_TYPE_Refreshed_MEAN,PREV_NAME_CLIENT_TYPE_Repeater_MEAN,PREV_NAME_CLIENT_TYPE_XNA_MEAN,PREV_NAME_CLIENT_TYPE_nan_MEAN,PREV_NAME_GOODS_CATEGORY_Additional Service_MEAN,PREV_NAME_GOODS_CATEGORY_Animals_MEAN,PREV_NAME_GOODS_CATEGORY_Audio/Video_MEAN,PREV_NAME_GOODS_CATEGORY_Auto Accessories_MEAN,PREV_NAME_GOODS_CATEGORY_Clothing and Accessories_MEAN,PREV_NAME_GOODS_CATEGORY_Computers_MEAN,PREV_NAME_GOODS_CATEGORY_Construction Materials_MEAN,PREV_NAME_GOODS_CATEGORY_Consumer Electronics_MEAN,PREV_NAME_GOODS_CATEGORY_Direct Sales_MEAN,PREV_NAME_GOODS_CATEGORY_Education_MEAN,PREV_NAME_GOODS_CATEGORY_Fitness_MEAN,PREV_NAME_GOODS_CATEGORY_Furniture_MEAN,PREV_NAME_GOODS_CATEGORY_Gardening_MEAN,PREV_NAME_GOODS_CATEGORY_Homewares_MEAN,PREV_NAME_GOODS_CATEGORY_House Construction_MEAN,PREV_NAME_GOODS_CATEGORY_Insurance_MEAN,PREV_NAME_GOODS_CATEGORY_Jewelry_MEAN,PREV_NAME_GOODS_CATEGORY_Medical Supplies_MEAN,PREV_NAME_GOODS_CATEGORY_Medicine_MEAN,PREV_NAME_GOODS_CATEGORY_Mobile_MEAN,PREV_NAME_GOODS_CATEGORY_Office Appliances_MEAN,PREV_NAME_GOODS_CATEGORY_Other_MEAN,PREV_NAME_GOODS_CATEGORY_Photo / Cinema Equipment_MEAN,PREV_NAME_GOODS_CATEGORY_Sport and Leisure_MEAN,PREV_NAME_GOODS_CATEGORY_Tourism_MEAN,PREV_NAME_GOODS_CATEGORY_Vehicles_MEAN,PREV_NAME_GOODS_CATEGORY_Weapon_MEAN,PREV_NAME_GOODS_CATEGORY_XNA_MEAN,PREV_NAME_GOODS_CATEGORY_nan_MEAN,PREV_NAME_PORTFOLIO_Cards_MEAN,PREV_NAME_PORTFOLIO_Cars_MEAN,PREV_NAME_PORTFOLIO_Cash_MEAN,PREV_NAME_PORTFOLIO_POS_MEAN,PREV_NAME_PORTFOLIO_XNA_MEAN,PREV_NAME_PORTFOLIO_nan_MEAN,PREV_NAME_PRODUCT_TYPE_XNA_MEAN,PREV_NAME_PRODUCT_TYPE_walk-in_MEAN,PREV_NAME_PRODUCT_TYPE_x-sell_MEAN,PREV_NAME_PRODUCT_TYPE_nan_MEAN,PREV_CHANNEL_TYPE_AP+ (Cash loan)_MEAN,PREV_CHANNEL_TYPE_Car dealer_MEAN,PREV_CHANNEL_TYPE_Channel of corporate sales_MEAN,PREV_CHANNEL_TYPE_Contact center_MEAN,PREV_CHANNEL_TYPE_Country-wide_MEAN,PREV_CHANNEL_TYPE_Credit and cash offices_MEAN,PREV_CHANNEL_TYPE_Regional / Local_MEAN,PREV_CHANNEL_TYPE_Stone_MEAN,PREV_CHANNEL_TYPE_nan_MEAN,PREV_NAME_SELLER_INDUSTRY_Auto technology_MEAN,PREV_NAME_SELLER_INDUSTRY_Clothing_MEAN,PREV_NAME_SELLER_INDUSTRY_Connectivity_MEAN,PREV_NAME_SELLER_INDUSTRY_Construction_MEAN,PREV_NAME_SELLER_INDUSTRY_Consumer electronics_MEAN,PREV_NAME_SELLER_INDUSTRY_Furniture_MEAN,PREV_NAME_SELLER_INDUSTRY_Industry_MEAN,PREV_NAME_SELLER_INDUSTRY_Jewelry_MEAN,PREV_NAME_SELLER_INDUSTRY_MLM partners_MEAN,PREV_NAME_SELLER_INDUSTRY_Tourism_MEAN,PREV_NAME_SELLER_INDUSTRY_XNA_MEAN,PREV_NAME_SELLER_INDUSTRY_nan_MEAN,PREV_NAME_YIELD_GROUP_XNA_MEAN,PREV_NAME_YIELD_GROUP_high_MEAN,PREV_NAME_YIELD_GROUP_low_action_MEAN,PREV_NAME_YIELD_GROUP_low_normal_MEAN,PREV_NAME_YIELD_GROUP_middle_MEAN,PREV_NAME_YIELD_GROUP_nan_MEAN,PREV_PRODUCT_COMBINATION_Card Street_MEAN,PREV_PRODUCT_COMBINATION_Card X-Sell_MEAN,PREV_PRODUCT_COMBINATION_Cash_MEAN,PREV_PRODUCT_COMBINATION_Cash Street: high_MEAN,PREV_PRODUCT_COMBINATION_Cash Street: low_MEAN,PREV_PRODUCT_COMBINATION_Cash Street: middle_MEAN,PREV_PRODUCT_COMBINATION_Cash X-Sell: high_MEAN,PREV_PRODUCT_COMBINATION_Cash X-Sell: low_MEAN,PREV_PRODUCT_COMBINATION_Cash X-Sell: middle_MEAN,PREV_PRODUCT_COMBINATION_POS household with interest_MEAN,PREV_PRODUCT_COMBINATION_POS household without interest_MEAN,PREV_PRODUCT_COMBINATION_POS industry with interest_MEAN,PREV_PRODUCT_COMBINATION_POS industry without interest_MEAN,PREV_PRODUCT_COMBINATION_POS mobile with interest_MEAN,PREV_PRODUCT_COMBINATION_POS mobile without interest_MEAN,PREV_PRODUCT_COMBINATION_POS other with interest_MEAN,PREV_PRODUCT_COMBINATION_POS others without interest_MEAN,PREV_PRODUCT_COMBINATION_nan_MEAN,APPROVED_AMT_ANNUITY_MIN,APPROVED_AMT_ANNUITY_MAX,APPROVED_AMT_ANNUITY_MEAN,APPROVED_AMT_APPLICATION_MIN,APPROVED_AMT_APPLICATION_MAX,APPROVED_AMT_APPLICATION_MEAN,APPROVED_AMT_CREDIT_MIN,APPROVED_AMT_CREDIT_MAX,APPROVED_AMT_CREDIT_MEAN,APPROVED_APP_CREDIT_PERC_MIN,APPROVED_APP_CREDIT_PERC_MAX,APPROVED_APP_CREDIT_PERC_MEAN,APPROVED_APP_CREDIT_PERC_VAR,APPROVED_AMT_DOWN_PAYMENT_MIN,APPROVED_AMT_DOWN_PAYMENT_MAX,APPROVED_AMT_DOWN_PAYMENT_MEAN,APPROVED_AMT_GOODS_PRICE_MIN,APPROVED_AMT_GOODS_PRICE_MAX,APPROVED_AMT_GOODS_PRICE_MEAN,APPROVED_HOUR_APPR_PROCESS_START_MIN,APPROVED_HOUR_APPR_PROCESS_START_MAX,APPROVED_HOUR_APPR_PROCESS_START_MEAN,APPROVED_RATE_DOWN_PAYMENT_MIN,APPROVED_RATE_DOWN_PAYMENT_MAX,APPROVED_RATE_DOWN_PAYMENT_MEAN,APPROVED_DAYS_DECISION_MIN,APPROVED_DAYS_DECISION_MAX,APPROVED_DAYS_DECISION_MEAN,APPROVED_CNT_PAYMENT_MEAN,APPROVED_CNT_PAYMENT_SUM,REFUSED_AMT_ANNUITY_MIN,REFUSED_AMT_ANNUITY_MAX,REFUSED_AMT_ANNUITY_MEAN,REFUSED_AMT_APPLICATION_MIN,REFUSED_AMT_APPLICATION_MAX,REFUSED_AMT_APPLICATION_MEAN,REFUSED_AMT_CREDIT_MIN,REFUSED_AMT_CREDIT_MAX,REFUSED_AMT_CREDIT_MEAN,REFUSED_APP_CREDIT_PERC_MIN,REFUSED_APP_CREDIT_PERC_MAX,REFUSED_APP_CREDIT_PERC_MEAN,REFUSED_APP_CREDIT_PERC_VAR,REFUSED_AMT_DOWN_PAYMENT_MIN,REFUSED_AMT_DOWN_PAYMENT_MAX,REFUSED_AMT_DOWN_PAYMENT_MEAN,REFUSED_AMT_GOODS_PRICE_MIN,REFUSED_AMT_GOODS_PRICE_MAX,REFUSED_AMT_GOODS_PRICE_MEAN,REFUSED_HOUR_APPR_PROCESS_START_MIN,REFUSED_HOUR_APPR_PROCESS_START_MAX,REFUSED_HOUR_APPR_PROCESS_START_MEAN,REFUSED_RATE_DOWN_PAYMENT_MIN,REFUSED_RATE_DOWN_PAYMENT_MAX,REFUSED_RATE_DOWN_PAYMENT_MEAN,REFUSED_DAYS_DECISION_MIN,REFUSED_DAYS_DECISION_MAX,REFUSED_DAYS_DECISION_MEAN,REFUSED_CNT_PAYMENT_MEAN,REFUSED_CNT_PAYMENT_SUM,POS_MONTHS_BALANCE_MAX,POS_MONTHS_BALANCE_MEAN,POS_MONTHS_BALANCE_SIZE,POS_SK_DPD_MAX,POS_SK_DPD_MEAN,POS_SK_DPD_DEF_MAX,POS_SK_DPD_DEF_MEAN,POS_NAME_CONTRACT_STATUS_Active_MEAN,POS_NAME_CONTRACT_STATUS_Amortized debt_MEAN,POS_NAME_CONTRACT_STATUS_Approved_MEAN,POS_NAME_CONTRACT_STATUS_Canceled_MEAN,POS_NAME_CONTRACT_STATUS_Completed_MEAN,POS_NAME_CONTRACT_STATUS_Demand_MEAN,POS_NAME_CONTRACT_STATUS_Returned to the store_MEAN,POS_NAME_CONTRACT_STATUS_Signed_MEAN,POS_NAME_CONTRACT_STATUS_XNA_MEAN,POS_NAME_CONTRACT_STATUS_nan_MEAN,POS_COUNT,INSTAL_NUM_INSTALMENT_VERSION_NUNIQUE,INSTAL_DPD_MAX,INSTAL_DPD_MEAN,INSTAL_DPD_SUM,INSTAL_DBD_MAX,INSTAL_DBD_MEAN,INSTAL_DBD_SUM,INSTAL_PAYMENT_PERC_MAX,INSTAL_PAYMENT_PERC_MEAN,INSTAL_PAYMENT_PERC_SUM,INSTAL_PAYMENT_PERC_VAR,INSTAL_PAYMENT_DIFF_MAX,INSTAL_PAYMENT_DIFF_MEAN,INSTAL_PAYMENT_DIFF_SUM,INSTAL_PAYMENT_DIFF_VAR,INSTAL_AMT_INSTALMENT_MAX,INSTAL_AMT_INSTALMENT_MEAN,INSTAL_AMT_INSTALMENT_SUM,INSTAL_AMT_PAYMENT_MIN,INSTAL_AMT_PAYMENT_MAX,INSTAL_AMT_PAYMENT_MEAN,INSTAL_AMT_PAYMENT_SUM,INSTAL_DAYS_ENTRY_PAYMENT_MAX,INSTAL_DAYS_ENTRY_PAYMENT_MEAN,INSTAL_DAYS_ENTRY_PAYMENT_SUM,INSTAL_COUNT,CC_MONTHS_BALANCE_MIN,CC_MONTHS_BALANCE_MAX,CC_MONTHS_BALANCE_MEAN,CC_MONTHS_BALANCE_SUM,CC_MONTHS_BALANCE_VAR,CC_AMT_BALANCE_MIN,CC_AMT_BALANCE_MAX,CC_AMT_BALANCE_MEAN,CC_AMT_BALANCE_SUM,CC_AMT_BALANCE_VAR,CC_AMT_CREDIT_LIMIT_ACTUAL_MIN,CC_AMT_CREDIT_LIMIT_ACTUAL_MAX,CC_AMT_CREDIT_LIMIT_ACTUAL_MEAN,CC_AMT_CREDIT_LIMIT_ACTUAL_SUM,CC_AMT_CREDIT_LIMIT_ACTUAL_VAR,CC_AMT_DRAWINGS_ATM_CURRENT_MIN,CC_AMT_DRAWINGS_ATM_CURRENT_MAX,CC_AMT_DRAWINGS_ATM_CURRENT_MEAN,CC_AMT_DRAWINGS_ATM_CURRENT_SUM,CC_AMT_DRAWINGS_ATM_CURRENT_VAR,CC_AMT_DRAWINGS_CURRENT_MIN,CC_AMT_DRAWINGS_CURRENT_MAX,CC_AMT_DRAWINGS_CURRENT_MEAN,CC_AMT_DRAWINGS_CURRENT_SUM,CC_AMT_DRAWINGS_CURRENT_VAR,CC_AMT_DRAWINGS_OTHER_CURRENT_MIN,CC_AMT_DRAWINGS_OTHER_CURRENT_MAX,CC_AMT_DRAWINGS_OTHER_CURRENT_MEAN,CC_AMT_DRAWINGS_OTHER_CURRENT_SUM,CC_AMT_DRAWINGS_OTHER_CURRENT_VAR,CC_AMT_DRAWINGS_POS_CURRENT_MIN,CC_AMT_DRAWINGS_POS_CURRENT_MAX,CC_AMT_DRAWINGS_POS_CURRENT_MEAN,CC_AMT_DRAWINGS_POS_CURRENT_SUM,CC_AMT_DRAWINGS_POS_CURRENT_VAR,CC_AMT_INST_MIN_REGULARITY_MIN,CC_AMT_INST_MIN_REGULARITY_MAX,CC_AMT_INST_MIN_REGULARITY_MEAN,CC_AMT_INST_MIN_REGULARITY_SUM,CC_AMT_INST_MIN_REGULARITY_VAR,CC_AMT_PAYMENT_CURRENT_MIN,CC_AMT_PAYMENT_CURRENT_MAX,CC_AMT_PAYMENT_CURRENT_MEAN,CC_AMT_PAYMENT_CURRENT_SUM,CC_AMT_PAYMENT_CURRENT_VAR,CC_AMT_PAYMENT_TOTAL_CURRENT_MIN,CC_AMT_PAYMENT_TOTAL_CURRENT_MAX,CC_AMT_PAYMENT_TOTAL_CURRENT_MEAN,CC_AMT_PAYMENT_TOTAL_CURRENT_SUM,CC_AMT_PAYMENT_TOTAL_CURRENT_VAR,CC_AMT_RECEIVABLE_PRINCIPAL_MIN,CC_AMT_RECEIVABLE_PRINCIPAL_MAX,CC_AMT_RECEIVABLE_PRINCIPAL_MEAN,CC_AMT_RECEIVABLE_PRINCIPAL_SUM,CC_AMT_RECEIVABLE_PRINCIPAL_VAR,CC_AMT_RECIVABLE_MIN,CC_AMT_RECIVABLE_MAX,CC_AMT_RECIVABLE_MEAN,CC_AMT_RECIVABLE_SUM,CC_AMT_RECIVABLE_VAR,CC_AMT_TOTAL_RECEIVABLE_MIN,CC_AMT_TOTAL_RECEIVABLE_MAX,CC_AMT_TOTAL_RECEIVABLE_MEAN,CC_AMT_TOTAL_RECEIVABLE_SUM,CC_AMT_TOTAL_RECEIVABLE_VAR,CC_CNT_DRAWINGS_ATM_CURRENT_MIN,CC_CNT_DRAWINGS_ATM_CURRENT_MAX,CC_CNT_DRAWINGS_ATM_CURRENT_MEAN,CC_CNT_DRAWINGS_ATM_CURRENT_SUM,CC_CNT_DRAWINGS_ATM_CURRENT_VAR,CC_CNT_DRAWINGS_CURRENT_MIN,CC_CNT_DRAWINGS_CURRENT_MAX,CC_CNT_DRAWINGS_CURRENT_MEAN,CC_CNT_DRAWINGS_CURRENT_SUM,CC_CNT_DRAWINGS_CURRENT_VAR,CC_CNT_DRAWINGS_OTHER_CURRENT_MIN,CC_CNT_DRAWINGS_OTHER_CURRENT_MAX,CC_CNT_DRAWINGS_OTHER_CURRENT_MEAN,CC_CNT_DRAWINGS_OTHER_CURRENT_SUM,CC_CNT_DRAWINGS_OTHER_CURRENT_VAR,CC_CNT_DRAWINGS_POS_CURRENT_MIN,CC_CNT_DRAWINGS_POS_CURRENT_MAX,CC_CNT_DRAWINGS_POS_CURRENT_MEAN,CC_CNT_DRAWINGS_POS_CURRENT_SUM,CC_CNT_DRAWINGS_POS_CURRENT_VAR,CC_CNT_INSTALMENT_MATURE_CUM_MIN,CC_CNT_INSTALMENT_MATURE_CUM_MAX,CC_CNT_INSTALMENT_MATURE_CUM_MEAN,CC_CNT_INSTALMENT_MATURE_CUM_SUM,CC_CNT_INSTALMENT_MATURE_CUM_VAR,CC_SK_DPD_MIN,CC_SK_DPD_MAX,CC_SK_DPD_MEAN,CC_SK_DPD_SUM,CC_SK_DPD_VAR,CC_SK_DPD_DEF_MIN,CC_SK_DPD_DEF_MAX,CC_SK_DPD_DEF_MEAN,CC_SK_DPD_DEF_SUM,CC_SK_DPD_DEF_VAR,CC_NAME_CONTRACT_STATUS_Active_MIN,CC_NAME_CONTRACT_STATUS_Active_MAX,CC_NAME_CONTRACT_STATUS_Active_MEAN,CC_NAME_CONTRACT_STATUS_Active_SUM,CC_NAME_CONTRACT_STATUS_Active_VAR,CC_NAME_CONTRACT_STATUS_Approved_MIN,CC_NAME_CONTRACT_STATUS_Approved_MAX,CC_NAME_CONTRACT_STATUS_Approved_MEAN,CC_NAME_CONTRACT_STATUS_Approved_SUM,CC_NAME_CONTRACT_STATUS_Approved_VAR,CC_NAME_CONTRACT_STATUS_Completed_MIN,CC_NAME_CONTRACT_STATUS_Completed_MAX,CC_NAME_CONTRACT_STATUS_Completed_MEAN,CC_NAME_CONTRACT_STATUS_Completed_SUM,CC_NAME_CONTRACT_STATUS_Completed_VAR,CC_NAME_CONTRACT_STATUS_Demand_MIN,CC_NAME_CONTRACT_STATUS_Demand_MAX,CC_NAME_CONTRACT_STATUS_Demand_MEAN,CC_NAME_CONTRACT_STATUS_Demand_SUM,CC_NAME_CONTRACT_STATUS_Demand_VAR,CC_NAME_CONTRACT_STATUS_Refused_MIN,CC_NAME_CONTRACT_STATUS_Refused_MAX,CC_NAME_CONTRACT_STATUS_Refused_MEAN,CC_NAME_CONTRACT_STATUS_Refused_SUM,CC_NAME_CONTRACT_STATUS_Refused_VAR,CC_NAME_CONTRACT_STATUS_Sent proposal_MIN,CC_NAME_CONTRACT_STATUS_Sent proposal_MAX,CC_NAME_CONTRACT_STATUS_Sent proposal_MEAN,CC_NAME_CONTRACT_STATUS_Sent proposal_SUM,CC_NAME_CONTRACT_STATUS_Sent proposal_VAR,CC_NAME_CONTRACT_STATUS_Signed_MIN,CC_NAME_CONTRACT_STATUS_Signed_MAX,CC_NAME_CONTRACT_STATUS_Signed_MEAN,CC_NAME_CONTRACT_STATUS_Signed_SUM,CC_NAME_CONTRACT_STATUS_Signed_VAR,CC_NAME_CONTRACT_STATUS_nan_MIN,CC_NAME_CONTRACT_STATUS_nan_MAX,CC_NAME_CONTRACT_STATUS_nan_MEAN,CC_NAME_CONTRACT_STATUS_nan_SUM,CC_NAME_CONTRACT_STATUS_nan_VAR,CC_COUNT,TARGET
+100002,0,0,0,0,202500.0,406597.5,24700.5,351000.0,0.018801,-9461,-637.0,-3648.0,-2120,,1,1,0,1,1,0,1.0,2,2,10,0,0,0,0,0,0,0.0830369673913225,0.2629485927471776,0.1393757800997895,0.0247,0.0369,0.9722,0.6192,0.0143,0.0,0.069,0.0833,0.125,0.0369,0.0202,0.019,0.0,0.0,0.0252,0.0383,0.9722,0.6341,0.0144,0.0,0.069,0.0833,0.125,0.0377,0.022,0.0198,0.0,0.0,0.025,0.0369,0.9722,0.6243,0.0144,0.0,0.069,0.0833,0.125,0.0375,0.0205,0.0193,0.0,0.0,0.0149,2.0,2.0,2.0,2.0,-1134.0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0.0,0.0,0.0,0.0,0.0,1.0,True,False,False,False,False,False,False,False,True,False,False,False,False,False,False,True,False,False,False,False,True,False,False,False,True,False,False,True,False,False,False,False,False,False,False,False,False,False,False,False,True,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,True,False,False,False,False,False,True,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,False,True,False,True,False,False,False,False,False,False,False,True,False,True,False,0.0673290349857309,0.4980355265342261,202500.0,0.12197777777777778,0.06074926678103038,-1437.0,-103.0,-874.0,186150.0,-1072.0,780.0,-349.0,-499.875,0.0,0.0,1681.029,450000.0,108131.945625,865055.565,245781.0,49156.2,245781.0,0.0,7997.14125,31988.565,0.0,0.0,0.0,-47.0,0.0,13.75,110.0,0.25,0.0,0.75,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.5,0.5,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.4069602272727273,0.2556818181818182,0.0,0.0,0.0,0.0,0.17542613636363635,0.16193181818181818,0.0,-1042.0,-103.0,-572.5,440860.5,780.0,780.0,780.0,-15.5,0.0,0.0,40.5,450000.0,240994.2825,481988.565,245781.0,122890.5,245781.0,0.0,15994.2825,31988.565,0.0,0.0,0.0,-34.0,0.0,10.0,20.0,-1437.0,-476.0,-974.5,123956.7,-1072.0,85.0,-574.8,-661.3333333333334,0.0,0.0,2091.16125,135000.0,63844.5,383067.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,-47.0,0.0,15.0,90.0,9251.775,9251.775,9251.775,179055.0,179055.0,179055.0,179055.0,179055.0,179055.0,1.0,1.0,1.0,,0.0,0.0,0.0,179055.0,179055.0,179055.0,9.0,9.0,9.0,0.0,0.0,0.0,-606.0,-606.0,-606.0,24.0,24.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,9251.775,9251.775,9251.775,179055.0,179055.0,179055.0,179055.0,179055.0,179055.0,1.0,1.0,1.0,,0.0,0.0,0.0,179055.0,179055.0,179055.0,9.0,9.0,9.0,0.0,0.0,0.0,-606.0,-606.0,-606.0,24.0,24.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,-1.0,-10.0,19.0,0.0,0.0,0.0,0.0,1.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,19.0,2.0,0.0,0.0,0.0,31.0,20.42105263157895,388.0,1.0,1.0,19.0,0.0,0.0,0.0,0.0,0.0,53093.745,11559.247105263159,219625.695,9251.775,53093.745,11559.247105263159,219625.695,-49.0,-315.42105263157896,-5993.0,19.0,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,1

3_Results/best_gradient_boosting_model.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb128d0274b1a5ad89a2d745acc391f39edf9115f557e0c6dc39c2612092dd29
+size 1906744

Dockerfile

@@ -0,0 +1,33 @@
+# Dockerfile pour Hugging Face Spaces
+FROM python:3.11-slim
+
+# Créer un utilisateur non-root (requis par HF Spaces)
+RUN useradd -m -u 1000 user
+
+# Définir le répertoire de travail
+WORKDIR /app
+
+# Copier les fichiers de dépendances
+COPY --chown=user requirements.txt .
+
+# Installer les dépendances
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r requirements.txt
+
+# Copier le code de l'application
+COPY --chown=user api.py .
+COPY --chown=user 2_Data_transformed/app_train_Enc_wo_Outliers_Feat_Eng_Join_Align_head.csv ./2_Data_transformed/
+COPY --chown=user 3_Results/best_gradient_boosting_model.pkl ./3_Results/
+
+# Changer vers l'utilisateur non-root
+USER user
+
+# Exposer le port 7860 (port par défaut de HF Spaces)
+EXPOSE 7860
+
+# Variables d'environnement
+ENV PYTHONUNBUFFERED=1
+
+# Commande de démarrage
+CMD ["uvicorn", "api:app", "--host", "0.0.0.0", "--port", "7860"]
+

README.md

@@ -1,12 +1,24 @@
 ---
-title: Oc Mlops Projet 2
-emoji: 🦀
-colorFrom: yellow
-colorTo: gray
+title: Prêt à Dépenser - API de Prédiction
+emoji: 💰
+colorFrom: blue
+colorTo: green
 sdk: docker
 pinned: false
-license: apache-2.0
-short_description: Openclassrooms - Confirmez vos compétences en MLOps - Part2
+license: mit
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# API de Prédiction de Crédit - Prêt à Dépenser
+
+Cette API permet de prédire si un client risque d'être en défaut de paiement.
+
+## Endpoints disponibles
+
+- `GET /health` - Vérification de l'état de l'API
+- `GET /columns` - Liste des colonnes attendues
+- `POST /predict` - Prédiction pour un client
+- `POST /predict/file` - Prédiction en lot via fichier CSV
+
+## Documentation interactive
+
+Accédez à la documentation Swagger : `/docs`

api.py

@@ -0,0 +1,330 @@
+"""
+API REST FastAPI pour les prédictions de modèle ML.
+
+Cette API charge un modèle pickle au démarrage et expose des endpoints
+pour effectuer des prédictions à partir de variables d'entrée.
+"""
+
+import logging
+import time
+import os
+import json
+from datetime import datetime, timezone
+from fastapi import FastAPI, HTTPException, UploadFile, File
+import io
+from pydantic import BaseModel
+import pickle
+from typing import Dict, Any, List
+import numpy as np
+import pandas as pd
+
+
+class JsonFormatter(logging.Formatter):
+    """
+    Formateur JSON pour les logs.
+    """
+    def format(self, record: logging.LogRecord) -> str:
+        log_record = {
+            "timestamp": datetime.now(timezone.utc).isoformat(),
+            "level": record.levelname,
+            "message": record.getMessage(),
+            "module": record.module,
+            "function": record.funcName,
+            "line": record.lineno
+        }
+        if record.exc_info:
+            log_record["exception"] = self.formatException(record.exc_info)
+        return json.dumps(log_record, ensure_ascii=False)
+
+
+# Configuration du logging avec handler JSON
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.INFO)
+
+json_formatter = JsonFormatter()
+
+# Chemin du fichier de log (utiliser /tmp pour HF Spaces où l'écriture est autorisée)
+LOG_FILE_PATH = os.environ.get("LOG_FILE_PATH", "/tmp/api_log.json" if os.path.exists("/tmp") else "./api_log.json")
+
+try:
+    file_handler = logging.FileHandler(LOG_FILE_PATH)
+    file_handler.setFormatter(json_formatter)
+    logger.addHandler(file_handler)
+except PermissionError:
+    pass  # Ignorer si on ne peut pas écrire le fichier de log
+
+stream_handler = logging.StreamHandler()
+stream_handler.setFormatter(json_formatter)
+
+logger.addHandler(stream_handler)
+
+# Initialisation de l'application FastAPI
+app = FastAPI(
+    title="API de Prédiction ML",
+    description="API pour effectuer des prédictions avec un modèle de Machine Learning",
+    version="1.0.0"
+)
+
+# Répertoire de base de l'application (pour compatibilité HF Spaces)
+BASE_DIR = os.path.dirname(os.path.abspath(__file__))
+
+# Chemin vers le fichier du modèle pickle
+MODEL_PATH = os.path.join(BASE_DIR, "3_Results/best_gradient_boosting_model.pkl")
+
+# Chemin vers le fichier CSV pour récupérer l'ordre des colonnes
+CSV_PATH = os.path.join(BASE_DIR, "2_Data_transformed/app_train_Enc_wo_Outliers_Feat_Eng_Join_Align_head.csv")
+
+# Chemin vers le fichier CSV pour enregistrer les données (détection de data drift)
+# Sur HF Spaces, utiliser /tmp pour les fichiers temporaires (écriture autorisée)
+DRIFT_LOG_PATH = os.environ.get("DRIFT_LOG_PATH", "/tmp/data_io.csv" if os.path.exists("/tmp") else os.path.join(BASE_DIR, "data_io.csv"))
+
+# Seuil de décision pour la classification (optimisé pour le métier)
+THRESHOLD = 0.474
+
+
+def log_data_for_drift(input_df: pd.DataFrame, predictions: list):
+    """
+    Enregistre les données d'entrée et de sortie pour la détection de data drift.
+
+    Args:
+        input_df: DataFrame contenant les features d'entrée.
+        predictions: Liste des prédictions effectuées.
+    """
+    try:
+        # Ajouter timestamp et prédictions
+        log_df = input_df.copy()
+        log_df['_timestamp'] = datetime.now().isoformat()
+        log_df['_prediction'] = predictions
+
+        # Vérifier si le fichier existe pour ajouter ou créer
+        file_exists = os.path.exists(DRIFT_LOG_PATH)
+
+        # Écrire dans le fichier CSV (mode append)
+        log_df.to_csv(
+            DRIFT_LOG_PATH,
+            mode='a',
+            header=not file_exists,
+            index=False,
+            sep=';'
+        )
+        logger.info(f"Données enregistrées pour drift detection: {len(log_df)} lignes")
+    except Exception as e:
+        logger.warning(f"Impossible d'enregistrer les données pour drift: {str(e)}")
+
+
+def load_model():
+    """
+    Charge le modèle ML depuis un fichier pickle.
+
+    Returns:
+        Le modèle chargé ou None si le fichier n'existe pas.
+    """
+    try:
+        with open(MODEL_PATH, "rb") as f:
+            model = pickle.load(f)
+            logger.info(f"Modèle chargé avec succès depuis {MODEL_PATH}")
+            return model
+    except FileNotFoundError:
+        logger.error(f"Fichier modèle non trouvé: {MODEL_PATH}")
+        return None
+
+
+def load_column_order() -> List[str]:
+    """
+    Charge le fichier CSV et extrait l'ordre des colonnes.
+
+    Returns:
+        Liste des noms de colonnes dans l'ordre du fichier CSV,
+        ou liste vide si le fichier n'existe pas.
+    """
+    try:
+        df = pd.read_csv(CSV_PATH, nrows=0)  # Charger uniquement les en-têtes
+        logger.info(f"Nombre et ordre des colonnes chargé depuis {CSV_PATH}")
+    except FileNotFoundError:
+        logger.error(f"Fichier CSV non trouvé: {CSV_PATH}")
+        return []
+    try:
+        df.drop(columns=['SK_ID_CURR', 'TARGET'], inplace=True)
+    except KeyError:
+        pass  # Si 'SK_ID_CURR', 'TARGET' ne sont pas présents, ignorer l'erreur
+    logger.info(f"Nombre de colonnes chargées: {len(df.columns)}")
+    return df.columns.tolist()
+
+# Chargement du modèle au démarrage de l'application
+model = load_model()
+
+# Chargement de l'ordre des colonnes au démarrage
+column_order = load_column_order()
+
+logger.info("API initialisée et prête")
+
+
+class PredictionInput(BaseModel):
+    """
+    Modèle Pydantic pour les données d'entrée de prédiction.
+
+    Attributes:
+        features: Dictionnaire contenant les noms des variables et leurs valeurs.
+    """
+    features: Dict[str, Any]
+
+
+class PredictionOutput(BaseModel):
+    """
+    Modèle Pydantic pour la réponse de prédiction.
+
+    Attributes:
+        prediction: Résultat de la prédiction du modèle (0=accepté, 1=rejeté).
+        probability: Probabilité de défaut (classe 1).
+        threshold: Seuil de décision utilisé.
+        status: Statut de la requête.
+    """
+    prediction: int
+    probability: float
+    threshold: float
+    status: str
+
+
+@app.post("/predict", response_model=PredictionOutput)
+def predict(input_data: PredictionInput):
+    """
+    Endpoint pour effectuer une prédiction.
+
+    Args:
+        input_data: Dictionnaire des features à utiliser pour la prédiction.
+
+    Returns:
+        PredictionOutput contenant la prédiction et le statut.
+    """
+    start_time = time.time()
+    logger.info("Requête de prédiction reçue")
+
+    if model is None:
+        logger.error("Tentative de prédiction sans modèle chargé")
+        raise HTTPException(status_code=500, detail="Modèle non chargé")
+
+    if not column_order:
+        logger.error("Ordre des colonnes non disponible")
+        raise HTTPException(status_code=500, detail="Ordre des colonnes non chargé")
+
+    try:
+        # Réordonner les features selon l'ordre des colonnes du CSV
+        feature_values = {col: [input_data.features.get(col, np.nan)] for col in column_order}
+        X = pd.DataFrame(feature_values)
+
+        # Exécuter la prédiction avec le modèle
+        probabilities = model.predict_proba(X)
+        proba_default = probabilities[0][1]  # Probabilité de la classe 1 (défaut)
+        prediction = 1 if proba_default >= THRESHOLD else 0
+
+        # Enregistrer les données pour la détection de drift
+        log_data_for_drift(X, [prediction])
+
+        execution_time = time.time() - start_time
+        logger.info(f"Prédiction effectuée avec succès: {prediction} (proba={proba_default:.4f}, seuil={THRESHOLD}, temps={execution_time:.4f}s)")
+
+        return PredictionOutput(
+            prediction=prediction,
+            probability=round(proba_default, 4),
+            threshold=THRESHOLD,
+            status="success"
+        )
+    except Exception as e:
+        execution_time = time.time() - start_time
+        logger.error(f"Erreur lors de la prédiction: {str(e)} (temps d'exécution: {execution_time:.4f}s)")
+        raise HTTPException(status_code=400, detail=str(e))
+
+@app.post("/predict/file")
+async def predict_from_file(file: UploadFile = File(...)):
+    """
+    Endpoint pour effectuer des prédictions à partir d'un fichier CSV uploadé.
+
+    Args:
+        file: Fichier CSV contenant les features (une ou plusieurs lignes).
+
+    Returns:
+        Dictionnaire avec les prédictions pour chaque ligne.
+    """
+    start_time = time.time()
+    logger.info(f"Fichier reçu pour prédiction: {file.filename}")
+
+    if model is None:
+        logger.error("Tentative de prédiction sans modèle chargé")
+        raise HTTPException(status_code=500, detail="Modèle non chargé")
+
+    if not column_order:
+        logger.error("Ordre des colonnes non disponible")
+        raise HTTPException(status_code=500, detail="Ordre des colonnes non chargé")
+
+    # Vérifier l'extension du fichier
+    if not file.filename.endswith('.csv'):
+        logger.warning(f"Format de fichier invalide: {file.filename}")
+        raise HTTPException(status_code=400, detail="Le fichier doit être au format CSV")
+
+    try:
+        # Lire le contenu du fichier
+        contents = await file.read()
+        df = pd.read_csv(io.StringIO(contents.decode('utf-8')))
+        logger.info(f"Fichier CSV lu avec succès: {len(df)} lignes")
+
+        # Vérifier et réordonner les colonnes selon l'ordre attendu
+        missing_cols = set(column_order) - set(df.columns)
+        if missing_cols:
+            logger.error(f"Colonnes manquantes dans le fichier: {list(missing_cols)}")
+            raise HTTPException(
+                status_code=400,
+                detail=f"Colonnes manquantes: {list(missing_cols)}"
+            )
+
+        # Sélectionner uniquement les colonnes nécessaires dans le bon ordre
+        X = df[column_order]
+
+        # Exécuter les prédictions avec le seuil personnalisé
+        probabilities = model.predict_proba(X)
+        proba_defaults = [p[1] for p in probabilities]  # Probabilité de la classe 1 (défaut)
+        predictions = [1 if p >= THRESHOLD else 0 for p in proba_defaults]
+
+        # Enregistrer les données pour la détection de drift
+        log_data_for_drift(X, predictions)
+
+        execution_time = time.time() - start_time
+        logger.info(f"Prédictions effectuées avec succès: {len(predictions)} résultats (temps d'exécution: {execution_time:.4f}s)")
+
+        return {
+            "predictions": predictions,
+            "probabilities": [round(p, 4) for p in proba_defaults],
+            "threshold": THRESHOLD,
+            "count": len(predictions),
+            "status": "success"
+        }
+    except HTTPException:
+        raise
+    except Exception as e:
+        execution_time = time.time() - start_time
+        logger.error(f"Erreur lors du traitement du fichier: {str(e)} (temps d'exécution: {execution_time:.4f}s)")
+        raise HTTPException(status_code=400, detail=str(e))
+
+@app.get("/health")
+def health_check():
+    """
+    Endpoint de vérification de l'état de santé de l'API.
+    """
+    logger.debug("Vérification de santé de l'API")
+    return {
+        "status": "ok",
+        "model_loaded": model is not None,
+        "columns_loaded": len(column_order) > 0,
+        "num_features": len(column_order)
+    }
+
+
+@app.get("/columns")
+def get_columns():
+    """
+    Endpoint pour récupérer la liste des colonnes attendues.
+
+    Returns:
+        Liste des colonnes dans l'ordre attendu par le modèle.
+    """
+    logger.debug("Liste des colonnes demandée")
+    return {"columns": column_order, "count": len(column_order)}

requirements.txt

@@ -0,0 +1,15 @@
+# API FastAPI - Dépendances
+fastapi>=0.104.0
+uvicorn[standard]>=0.24.0
+pydantic>=2.0.0
+python-multipart>=0.0.6
+
+# Data Science
+numpy>=1.24.0
+pandas>=2.0.0
+scikit-learn>=1.3.0
+imbalanced-learn>=0.11.0
+
+# Modèle ML
+pickle5>=0.0.12;python_version<"3.8"
+