说明

自己第一次参加Kaggle比赛,Sber Russian Housing Market项目的练习代码

数据说明

用到的数据文件有train.csv、macro.csv和test.csv,分别包含了训练数据、对应的俄罗斯宏观经济数据和测试时用到的数据。

train.csv和test.csv中数据项含义

price_doc: sale price (this is the target variable),出售价格
id: transaction id,交易ID
timestamp: date of transaction, 交易日期
full_sq: total area in square meters, including loggias, balconies and other non-residential areas,总面积(平方米),包含走廊、阳台和其他公摊面积
life_sq: living area in square meters, excluding loggias, balconies and other non-residential areas, 使用面积,不包含走廊、阳台和其他公摊面积
floor: for apartments, floor of the building,公寓所在楼层
max_floor: number of floors in the building,总楼层
material: wall material,墙体材料
build_year: year built,修建年份
num_room: number of living rooms,客厅数目
kitch_sq: kitchen area,厨房面积
state: apartment condition,公寓条件
product_type: owner-occupier purchase or investment
sub_area: name of the district

The dataset also includes a collection of features about each property's surrounding neighbourhood, and some features that are constant across each sub area (known as a Raion). Most of the feature names are self explanatory, with the following notes.

full_all: subarea population
male_f, female_f: subarea population by gender
young_: population younger than working age
work_
: working-age population
ekder_: retirement-age population
n_m_{all|male|female}: population between n and m years old
build_count_
: buildings in the subarea by construction type or year
x_count_500: the number of x within 500m of the property
x_part_500: the share of x within 500m of the property
sqm: square meters
cafe_count_d_price_p: number of cafes within d meters of the property that have an average bill under p RUB
trc_: shopping malls
prom_: industrial zones
green_: green zones
metro_: subway
avto: distances by car
mkad_: Moscow Circle Auto Road
ttk_: Third Transport Ring
sadovoe_: Garden Ring
bulvar_ring_: Boulevard Ring
kremlin_: City center
zd_vokzaly_: Train station
oil_chemistry_: Dirty industry
ts_: Power plant

macro.csv中数据项含义

A set of macroeconomic indicators, one for each date.

timestamp: Transaction timestamp
oil_urals: Crude Oil Urals ($/bbl)
gdp_quart: GDP
gdp_quart_growth: Real GDP growth
cpi: Inflation - Consumer Price Index Growth
ppi: Inflation - Producer Price index Growth
gdp_deflator: Inflation - GDP deflator
balance_trade: Trade surplus
balance_trade_growth: Trade balance (as a percentage of previous year)
usdrub: Ruble/USD exchange rate
eurrub: Ruble/EUR exchange rate
brent: London Brent ($/bbl)
net_capital_export: Net import / export of capital
gdp_annual: GDP at current prices
gdp_annual_growth: GDP growth (in real terms)
average_provision_of_build_contract: Provision by orders in Russia (for the developer)
average_provision_of_build_contract_moscow: Provision by orders in Moscow (for the developer)
rts: Index RTS / return
micex: MICEX index / return
micex_rgbi_tr: MICEX index for government bonds (MICEX RGBI TR) / yield
micex_cbi_tr: MICEX Index corporate bonds (MICEX CBI TR) / yield
deposits_value: Volume of household deposits
deposits_growth: Volume growth of population's deposits
deposits_rate: Average interest rate on deposits
mortgage_value: Volume of mortgage loans
mortgage_growth: Growth of mortgage lending
mortgage_rate: Weighted average rate of mortgage loans
grp: GRP of the subject of Russian Federation where Apartment is located
grp_growth: Growth of gross regional product of the subject of the Russian Federation where Apartment is located
income_per_cap: Average income per capita
real_dispos_income_per_cap_growth: Growth in real disposable income of Population
salary: Average monthly salary
salary_growth: Growth of nominal wages
fixed_basket: Cost of a fixed basket of consumer goods and services for inter-regional comparisons of purchasing power
retail_trade_turnover: Retail trade turnover
retail_trade_turnover_per_cap: Retail trade turnover per capita
retail_trade_turnover_growth: Retail turnover (in comparable prices in% to corresponding period of previous year)
labor_force: Size of labor force
unemployment: Unemployment rate
employment: Employment rate
invest_fixed_capital_per_cap: Investments in fixed capital per capita
invest_fixed_assets: Absolute volume of investments in fixed assets
profitable_enterpr_share: Share of profitable enterprises
unprofitable_enterpr_share: The share of unprofitable enterprises
share_own_revenues: The share of own revenues in the total consolidated budget revenues
overdue_wages_per_cap: Overdue wages per person
fin_res_per_cap: The financial results of companies per capita
marriages_per_1000_cap: Number of marriages per 1,000 people
divorce_rate: The divorce rate / growth rate
construction_value: Volume of construction work performed (million rubles)
invest_fixed_assets_phys: The index of physical volume of investment in fixed assets (in comparable prices in% to the corresponding month of Previous year)
pop_natural_increase: Rate of natural increase / decrease in Population (1,000 persons)
pop_migration: Migration increase (decrease) of population
pop_total_inc: Total population growth
childbirth: Childbirth
mortality: Mortality
housing_fund_sqm: Housing Fund (sqm)
lodging_sqm_per_cap: Lodging (sqm / pax)
water_pipes_share: Plumbing availability (pax)
baths_share: Bath availability (pax)
sewerage_share: Canalization availability
gas_share: Gas (mains, liquefied) availability
hot_water_share: Hot water availability
electric_stove_share: Electric heating for the floor
heating_share: Heating availability
old_house_share: Proportion of old and dilapidated housing, percent
average_life_exp: Average life expectancy
infant_mortarity_per_1000_cap: Infant mortality rate (per 1,000 children aged up to one year)
perinatal_mort_per_1000_cap: Perinatal mortality rate (per 1,000 live births)
incidence_population: Overall incidence of the total population
rent_price_4+room_bus: rent price for 4-room apartment, business class
rent_price_3room_bus: rent price for 3-room apartment, business class
rent_price_2room_bus: rent price for 2-room apartment, business class
rent_price_1room_bus: rent price for 1-room apartment, business class
rent_price_3room_eco: rent price for 3-room apartment, econom class rent_price_2room_eco: rent price for 2-room apartment, econom class
rent_price_1room_eco: rent price for 1-room apartment, econom class
load_of_teachers_preschool_per_teacher: Load of teachers of preschool educational institutions (number of children per 100 teachers);
child_on_acc_pre_school: Number of children waiting for the determination to pre-school educational institutions, for capacity of 100
load_of_teachers_school_per_teacher: Load on teachers in high school (number of pupils in hugh school for 100 teachers)
students_state_oneshift: Proportion of pupils in high schools with one shift, of the total number of pupils in high schools
modern_education_share: Share of state (municipal) educational organizations, corresponding to modern requirements of education in the total number of high schools;
old_education_build_share: The share of state (municipal) educational organizations, buildings are in disrepair and in need of major repairs of the total number.
provision_doctors: Provision (relative number) of medical doctors in area
provision_nurse: Provision of nursing staff
load_on_doctors: The load on doctors (number of visits per physician)
power_clinics: Capacity of outpatient clinics
hospital_beds_available_per_cap: Availability of hospital beds per 100 000 persons
hospital_bed_occupancy_per_year: Average occupancy rate of the hospital beds during a year
provision_retail_space_sqm: Retail space
provision_retail_space_modern_sqm: Provision of population with retail space of modern formats, square meter
retail_trade_turnover_per_cap: Retail trade turnover per capita
turnover_catering_per_cap: Turnover of catering industry per person
theaters_viewers_per_1000_cap: Number of theaters viewers per 1000 population
seats_theather_rfmin_per_100000_cap: Total number of seats in Auditorium of the Ministry of Culture Russian theaters per 100,000 population
museum_visitis_per_100_cap: Number of visits to museums per 1000 of population
bandwidth_sports: Capacity of sports facilities
population_reg_sports_share: Proportion of population regularly doing sports
students_reg_sports_share: Proportion of pupils and students regularly doing sports in the total number
apartment_build: City residential apartment construction
apartment_fund_sqm: City residential apartment fund