-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathnate_stealing_kaitlyn_code.py
213 lines (185 loc) · 9.92 KB
/
nate_stealing_kaitlyn_code.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
# %% codecell
import pandas as pd
import seaborn as sns
import plotly.express as px
import plotly.offline as py
import matplotlib.pyplot as plt
import numpy as np
import pathlib
datafolder = pathlib.Path('Users\999Na\Documents\CarMax_Analytics')
data = datafolder / 'CaseCompetitionData' /'CaseCompetitionData.csv'
data = pd.read_csv(data, sep=',')
data['actual'] = 1
data.loc[data['subsequent_purchases'] == 0, 'actual'] = 0
data['customer_age'] = data['customer_age'].map({'21 - 30':25, '61 - 70':65, '31 - 40':35, '41 - 50':45, '51 - 60':55, '101+':102, '71 - 80':75,
'0 - 20':10, '81 - 90':85, '?':'?', '91 - 100':95})
data['customer_income'] = data['customer_income'].map({'20001 - 40000':30000, '80001 - 100000':90000, '40001 - 60000':50000, '60001 - 80000':70000,
'120001 - 140000':130000, '0 - 20000':10000, '100001 - 120000':110000, '160001 - 180000':170000,
'200001+':220000, '140001 - 160000':150000, '180001 - 200000':190000, '?':'?'})
data['purchase_price'] = data['purchase_price'].map({'10001 - 15000':12500, '15001 - 20000':17500,
'20001 - 25000':22500, '25001 - 30000':27500, '30001 - 35000':32500, '0 - 5000':2500, '45001 - 50000':47500,
'35001 - 40000':27500, '5001 - 10000':7500, '40001 - 45000':42500, '55001 - 60000':57500, '50001 - 55000':52500,
'60001 - 65000':62500, '65001 - 70000':67500, '70001 - 75000':72500, '75001 - 80000':77500, '85001 - 90000':87500,
'90001 - 95000':92500, '80001 - 85000':82500, '?':'?'})
data['counter'] = 1
price = data.loc[data['purchase_price'] != '?']
income = data.loc[data['customer_income'] != '?']
distance = data.loc[data['customer_distance_to_dealer'] != '?']
age = data.loc[data['customer_age'] != '?']
print(len(price))
print(len(income))
print(len(distance))
print(len(age))
distance['customer_distance_to_dealer'] = distance['customer_distance_to_dealer'].astype(np.int64)
# %% codecell
### Quantifying subs ratio for binary features
trade_in = data.groupby('trade_in')['actual'].mean().to_frame().reset_index()
print(trade_in.head())
customer_previous_purchase = data.groupby('customer_previous_purchase')['actual'].mean().to_frame().reset_index()
print(customer_previous_purchase.head())
warranty_used = data.groupby('vehicle_warranty_used')['actual'].mean().to_frame().reset_index()
print(warranty_used.head())
financing = data.groupby('vehicle_financing')['actual'].mean().to_frame().reset_index()
print(financing.head())
trade_in.to_csv('trade_subs.csv')
customer_previous_purchase.to_csv('prev_subs.csv')
warranty_used.to_csv('war_subs.csv')
financing.to_csv('fin_subs.csv')
# %% codecell
### Trade_in subs split distribution
par_cats = data[['actual', 'trade_in']]
par_cats.dtypes
fig = px.parallel_categories(par_cats)
py.offline.plot(fig, filename='parallel_cats.html', auto_open=True)
# %% codecell
purch_subs = data.groupby()
# %% codecell
### Trade_in purchase year subs ratio
purchase_year = data.loc[data['purchase_vehicle_year'] >= 2000].groupby(['trade_in', 'purchase_vehicle_year'])['actual'].mean().to_frame().reset_index()
purchase_year
fig = px.bar(purchase_year, x='purchase_vehicle_year', y='actual', facet_row='trade_in')
py.offline.plot(fig, 'year_previous.html', auto_open=True)
#purchase_year.to_csv('purchase_year.csv')
# %% codecell
### Trade in purchase year subs split distribution
par_cats = price.loc[price['purchase_vehicle_year'] >= 2000][['actual', 'purchase_vehicle_year', 'trade_in']]
par_cats.to_csv('trade_year_subs.csv')
#print(par_cats['purchase_vehicle_year'].unique())
#fig = px.parallel_categories(par_cats)
#py.offline.plot(fig, filename='parallel_cats.html', auto_open=True)
# %% codecell
### Trade in income subs ratio
income_trade = income.groupby(['customer_income', 'trade_in'])['actual'].mean().to_frame().reset_index()
#purchase_year
#fig = px.bar(income_trade, x='customer_income', y='actual', facet_row='trade_in')
#py.offline.plot(fig, 'income_trade.html', auto_open=True)
income_trade.to_csv('income_trade_subs.csv')
# %% codecell
### Trade in purchase price subs ratio
price_trade = price.groupby(['purchase_price', 'trade_in'])['actual'].mean().to_frame().reset_index()
#purchase_year
#fig = px.bar(price_trade, x='purchase_price', y='actual', facet_row='trade_in')
#py.offline.plot(fig, 'price_trade.html', auto_open=True)
#data['actual'].mean()
price_trade.to_csv('price_trade_subs.csv')
# %% codecell
### trade in age subs ratio
age_trade = age.groupby(['customer_age', 'trade_in'])['actual'].mean().to_frame().reset_index()
purchase_year
fig = px.bar(age_trade, x='customer_age', y='actual', facet_row='trade_in')
py.offline.plot(fig, 'age_trade.html', auto_open=True)
# %% codecell
### trade in distance to dealer subs ratio
dist_subs = distance.loc[distance['customer_distance_to_dealer'] <= 50].groupby(['customer_distance_to_dealer', 'customer_previous_purchase'])['actual'].mean().to_frame().reset_index()
purchase_year
fig = px.line(dist_subs, x='customer_distance_to_dealer', y='actual', facet_row='customer_previous_purchase')
py.offline.plot(fig, 'dist_trade.html', auto_open=True)
dist_subs.to_csv('dist_prev_subs_50.csv')
# %% codecell
### trade in make subs ratio
make_trade = data.groupby(['purchase_make', 'trade_in'])['actual'].mean().to_frame().reset_index()
#fig = px.bar(make_trade, x='purchase_make', y='actual', facet_row='trade_in')
#py.offline.plot(fig, 'make_trade.html', auto_open=True)
make_subs = data.groupby('purchase_make')['actual'].mean().to_frame().reset_index()
fig = px.bar(make_subs, x='purchase_make', y='actual')
py.offline.plot(fig, 'make_subs.html', auto_open=True)
# %% codecell
make_prev = data.groupby(['purchase_make', 'customer_previous_purchase'])['actual'].mean().to_frame().reset_index()
fig = px.bar(make_prev, x='purchase_make', y='actual', facet_row='customer_previous_purchase')
py.offline.plot(fig, 'make_prev.html', auto_open=True)
# %% codecell
makes = data
makes['counter'] = 1
makes = makes.groupby('purchase_make')['counter'].sum().to_frame().reset_index()
low_demand = makes.loc[makes['counter'] < 2000]
print(data.loc[data.purchase_make.isin(low_demand['purchase_make'].unique())]['actual'].mean())
low_make = data.loc[data.purchase_make.isin(low_demand['purchase_make'].unique())].groupby('purchase_make')['actual'].mean().to_frame().reset_index()
sns.barplot(x='purchase_make', y='actual', data=low_make)
print(low_make['actual'].mean())
low_make.to_csv('low_make_subs.csv')
#makes
low_demand.to_csv('low_demand_dist.csv')
make_subs.groupby('purchase_make')['actual'].mean().to_frame().reset_index().to_csv('make_subs.csv')
# %% codecell
make_fin = data.groupby(['purchase_make', 'vehicle_financing'])['actual'].mean().to_frame().reset_index()
#fig = px.bar(make_fin, x='purchase_make', y='actual', facet_row='vehicle_financing')
#py.offline.plot(fig, 'make_fin.html', auto_open=True)
make_fin.to_csv('make_fin_subs.csv')
# %% codecell
### Purchase price subs ratio
purch_ratio = price.groupby('purchase_price')['actual'].mean().to_frame().reset_index()
purch_ratio.to_csv('purch_subs.csv')
#fig = px.bar(purch_ratio, x='purchase_price', y='actual')
#py.offline.plot(fig, 'price_subs.html', auto_open=True)
# %% codecell
### purchase price subs=1 distribution
purch = price.loc[data['actual'] == 1].groupby('purchase_price')['actual'].sum().to_frame().reset_index()
fig = px.bar(purch, x='purchase_price', y='actual')
py.offline.plot(fig, 'price_subs_1_dist.html', auto_open=True)
purch
# %% codecell
### purchase price distribution
purch_dist = price.groupby('purchase_price')['insert_num'].count().to_frame().reset_index()
fig = px.bar(purch_dist, x='purchase_price', y='insert_num')
py.offline.plot(fig, 'price_dist.html', auto_open=True)
# %% codecell
### purchase price warranty subs ratio
warranty_price = price.groupby(['vehicle_warranty_used', 'purchase_price'])['actual'].mean().to_frame().reset_index()
fig = px.bar(warranty_price, x='purchase_price', y='actual', facet_row='vehicle_warranty_used')
py.offline.plot(fig, 'warranty_price.html', auto_open=True)
# %% codecell
### purchase price year subs ratio
price_year = price.loc[price['purchase_vehicle_year'] > 2000].groupby(['purchase_vehicle_year', 'purchase_price'])['actual'].mean().to_frame().reset_index()
fig = px.bar(price_year, x='purchase_price', y='actual', facet_row='purchase_vehicle_year', width=1600, height=1600)
py.offline.plot(fig, 'price_year.html', auto_open=True)
# %% codecell
### purchase price subs split distribution
par_cats = price[['actual', 'purchase_price']]
par_cats.dtypes
fig = px.parallel_categories(par_cats)
py.offline.plot(fig, filename='parallel_cats.html', auto_open=True)
# %% codecell
### purchase price income subs ratio
price_income = price.merge(income['insert_num'], on='insert_num')
price_income = price_income.groupby(['purchase_price', 'customer_income'])['actual'].mean().to_frame().reset_index()
fig = px.bar(price_income, x='purchase_price', y='actual', facet_row='customer_income', width=1600, height=1600)
py.offline.plot(fig, 'price_income.html', auto_open=True)
# %% codecell
### purchase price income distribution
price_income = price.merge(income['insert_num'], on='insert_num')
price_income = price_income.groupby(['purchase_price', 'customer_income'])['actual'].count().to_frame().reset_index()
fig = px.bar(price_income, x='purchase_price', y='actual', facet_row='customer_income', width=1600, height=1600)
py.offline.plot(fig, 'price_income.html', auto_open=True)
# %% codecell
stuff = data.loc[data['vehicle_financing'] == 0].groupby('customer_income')['insert_num'].count().to_frame().reset_index()
stuff = stuff.sort_values(by=['customer_income'])
sns.barplot(x='customer_income', y='insert_num', data=stuff)
# %% codecell
fin_in = income.groupby('vehicle_financing')['insert_num'].count().to_frame().reset_index()
fin_in.to_csv('fin_in.csv')
no_fin_in = income.loc[income['vehicle_financing'] == 0].groupby('customer_income')['insert_num'].count().to_frame().reset_index()
no_fin_in.to_csv('no_fin_in.csv')
#sns.barplot(x='customer_income', y='insert_num', data=no_fin_in)
# %% codecell
stuff = data.drop(columns=['subsequent_purchases', 'actual', 'insert_num'])
sns.heatmap(stuff.corr(), cmap="YlGnBu")