Kiva／クラウドファンディングの資金調達額予測

マルチモーダルコンペティション！Ridgelinezとの共同開催です

賞金: 200,000 円 参加ユーザー数: 231 3年弱前に終了
スポンサー
Ridgelinez株式会社
Aws4 request&x amz signedheaders=host&x amz signature=aa61ba011dec4a8d9defe40f25f8c5e31b4be58dde73af424b00fa244f3beccc

yshr10ic


folder_special 0
EDA

Simple EDA（CSVデータのみ）

EDA

  • train.csvとtest.csvを用いて簡単なEDAを実施してみました
  • Google Colaboratoryでの実行を想定しておりますが、複雑な処理はしておりませんので、必要なライブラリをインストールすればローカルでも問題なく動作すると思います
  • ご意見、ノートブックの不備などございましたら、コメントいただけると幸いです

事前準備

# mount drive

from google.colab import drive
drive.mount('/content/drive')
Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).

!pip install matplotlib-venn > /dev/null

設定

# カレントディレクトリを変更
import os
os.chdir('/content/drive/My Drive/Colab Notebooks/probspace/kiva/')
print(os.getcwd())
/content/drive/My Drive/Colab Notebooks/probspace/kiva

class Config():
    root_path = './'
    input_path = os.path.join(root_path, 'data')
import matplotlib.pylab as plt
from matplotlib_venn import venn2
import numpy as np
import pandas as pd

pd.set_option('max_columns', None)
pd.options.display.float_format = '{:.5f}'.format

データの読み込み

train_df = pd.read_csv(os.path.join(Config.input_path, 'train.csv'))
print(train_df.shape)
display(train_df)
(91333, 18)
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED LOAN_AMOUNT IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL
0 1733169 English Teodora is a 50-year-old married woman from th... Teodora is a 50-year-old married woman from th... 100 3115271 Weaving Arts to purchase materials like nipa palm, bamboo ... PH Philippines Maribojoc, Bohol shared 0.10000 PHP #Elderly monthly field_partner
1 1546998 English Diego is 32 years old and lives in the municip... Diego is 32 years old and lives in the municip... 1350 2870403 Barber Shop Services to buy two hair clippers, a new barber chair, ... CO Colombia Apartadó shared 0.10000 COP user_favorite, user_favorite monthly field_partner
2 1808517 Spanish Osman, es un joven de 27 años de edad, soltero... Osman is a young man, 27 years old, single, an... 225 3215705 Farming Agriculture to purchase sacks of fertilizers to care for a... HN Honduras Nueva Frontera, Santa Barbara. shared 0.10000 HNL NaN bullet field_partner
3 1452940 English His name is Nino, 31 years old, married to Che... His name is Nino, 31 years old, married to Che... 350 2745031 Motorcycle Transport Transportation to pay for fuel, tires and change oil for his ... PH Philippines Silang, Cavite shared 0.10000 PHP user_favorite monthly field_partner
4 1778420 English Pictured above is Teresa, often described as a... Pictured above is Teresa, often described as a... 625 3083800 Farming Agriculture to purchase hybrid seeds and fertilizer to imp... KE Kenya Mumias shared 0.10000 KES #Eco-friendly, #Sustainable Ag, #Parent, #Elde... bullet field_partner
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
91328 1688789 Spanish Rider tiene 20 años de edad, vive en San Javie... Rider is 20 years old. He lives in San Javier,... 775 3054018 Poultry Agriculture to buy chickens to raise and sell. EC Ecuador San Javier shared 0.10000 USD volunteer_like, #Animals, #Supporting Family monthly field_partner
91329 1878119 English Carmelita works hard to support four children.... Carmelita works hard to support four children.... 100 3311100 Personal Housing Expenses Housing to build a sanitary toilet for her family PH Philippines Danao Cebu standard nan PHP volunteer_like monthly field_partner
91330 1639680 English Orn, 60 years of age, appears in the photo. Sh... Orn, 60 years of age, appears in the photo. Sh... 1500 2990352 Grocery Store Food to pay for additional groceries to stock the s... KH Cambodia Takeo province shared 0.10000 USD user_favorite, #Elderly, user_favorite monthly field_partner
91331 1495391 Spanish Walter, a sus 27 años de edad, vive en unión l... At 27 years of age, Walter is in a live-in rel... 1750 2805390 Farming Agriculture to buy agricultural supplies, such as fertiliz... CO Colombia El Carmen de Viboral shared 0.10000 COP #Sustainable Ag, #Eco-friendly, user_favorite monthly field_partner
91332 1602898 English Greetings from Uganda! This is Godfrey. He is ... Greetings from Uganda! This is Godfrey. He is ... 275 2943724 Education provider Education to purchase a water-filtration system to provi... UG Uganda Isingiro shared 0.10000 UGX #Health and Sanitation, user_favorite, #School... irregular field_partner

91333 rows × 18 columns

test_df = pd.read_csv(os.path.join(Config.input_path, 'test.csv'))
print(test_df.shape)
display(test_df)
(91822, 17)
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL
0 2041445 English Marcela is 69 years old and married with ten c... Marcela is 69 years old and married with ten c... 4051101 General Store Retail to buy items to sell like canned goods and per... PH Philippines Cauayan, Negros Occidental standard nan PHP NaN monthly field_partner
1 1944435 English Roselia is 48 years old and has five children.... Roselia is 48 years old and has five children.... 3410523 Pigs Agriculture to buy feeds and other supplies to raise her pig PH Philippines Guihulngan, Negros Oriental standard nan PHP #Animals, #Repeat Borrower, #Schooling, #Woman... monthly field_partner
2 2083354 English Ma. Marebil is a single woman, 40 years old wi... Ma. Marebil is a single woman, 40 years old wi... 4146690 Clothing Sales Clothing to buy additional stock of clothes and dresses... PH Philippines Santa Barbara, Iloilo standard nan PHP #Parent, #Single Parent, #Woman-Owned Business monthly field_partner
3 1993565 English Good day, lenders! Meet one of KBMI’s clients,... Good day, lenders! Meet one of KBMI’s clients,... 3945982 Food Food to buy more foods to grow her business. ID Indonesia Pandeglang shared 0.10000 IDR #Woman-Owned Business, #Schooling, #Elderly, #... monthly field_partner
4 2064272 English Rosemarie is a married woman with two children... Rosemarie is a married woman with two children... 4114040 Food Food to buy ingredients for her food production bus... PH Philippines Sogod Cebu standard nan PHP NaN monthly field_partner
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
91817 1993862 English Marjorie is a resident of Tubigon, Bohol. She ... Marjorie is a resident of Tubigon, Bohol. She ... 3946629 Fishing Food to buy fishing nets. PH Philippines Tubigon, Bohol shared 0.00000 PHP #Parent, #Biz Durable Asset monthly field_partner
91818 2015070 English Hello, Kiva community! Meet Janeth, a mother e... Hello, Kiva community! Meet Janeth, a mother e... 4006025 Home Energy Personal Use to buy a solar lantern to provide adequate lig... KE Kenya Nandi Hills shared 0.00000 KES #Technology, #Eco-friendly, #Parent monthly field_partner
91819 1950349 French Agé de 32 ans, Komi est marié .C'est un bouche... Komi is 32 years old and married. He is a reno... 3423123 Butcher Shop Food to buy two cows. TG Togo Vakpossito shared 0.00000 XOF #Biz Durable Asset, user_favorite, #Animals monthly field_partner
91820 1921580 Russian Калбубу, 56 лет, вдова, есть взрослые дети. У ... Kalbubu is 56 years old, a widow, and she has ... 3373358 Dairy Agriculture to buy dairy cows to increase her headcount of... KG Kyrgyzstan Min-Bulak village, Talas region shared 0.00000 KGS #Animals, #Widowed, #Biz Durable Asset, #Woman... irregular field_partner
91821 1976733 English Hinrilyn is 31 years old and has two children.... Hinrilyn is 31 years old and has two children.... 3841884 Pigs Agriculture to buy feeds and other supplies to raise her l... PH Philippines Coron, Palawan standard nan PHP user_favorite, #Animals, #Parent, user_favorit... monthly field_partner

91822 rows × 17 columns

EDA

check null

display(
    pd.DataFrame({
        'train': train_df.isnull().sum(),
        'test': test_df.isnull().sum()
    }).reset_index()
)
index train test
0 ACTIVITY_NAME 0 0.00000
1 COUNTRY_CODE 0 0.00000
2 COUNTRY_NAME 0 0.00000
3 CURRENCY 0 0.00000
4 CURRENCY_EXCHANGE_COVERAGE_RATE 9272 43561.00000
5 CURRENCY_POLICY 0 0.00000
6 DESCRIPTION 0 0.00000
7 DESCRIPTION_TRANSLATED 1 0.00000
8 DISTRIBUTION_MODEL 0 0.00000
9 IMAGE_ID 0 0.00000
10 LOAN_AMOUNT 0 nan
11 LOAN_ID 0 0.00000
12 LOAN_USE 0 0.00000
13 ORIGINAL_LANGUAGE 0 0.00000
14 REPAYMENT_INTERVAL 0 0.00000
15 SECTOR_NAME 0 0.00000
16 TAGS 17986 30518.00000
17 TOWN_NAME 2760 2423.00000

check duplicate

print(f"[tr] {train_df.duplicated().sum()}")
print(f"[te] {test_df.duplicated().sum()}")
[tr] 0
[te] 0

all_df = pd.concat([train_df, test_df])
all_df.drop(columns=['LOAN_AMOUNT'], inplace=True)

print(f"{all_df.duplicated().sum()}")
0

LOAN_ID

  • LOAN_IDの重複はない
  • trainとtestでLOAN_IDはかぶっていない
print(f"[tr] len: {len(train_df)}, nunique: {train_df['LOAN_ID'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['LOAN_ID'].nunique()}")
[tr] len: 91333, nunique: 91333
[te] len: 91822, nunique: 91822

venn2([set(train_df['LOAN_ID']), set(test_df['LOAN_ID'])], set_labels=('train', 'test'));

ORIGINAL_LANGUAGE

  • 7割程度がEnglish
  • trainとtestで割合はだいたい同じ
display(
    pd.DataFrame({
        'train': train_df['ORIGINAL_LANGUAGE'].value_counts() / len(train_df),
        'test': test_df['ORIGINAL_LANGUAGE'].value_counts() / len(test_df)
    }).reset_index()
)
index train test
0 English 0.70005 0.70159
1 Spanish 0.20766 0.17450
2 French 0.05667 0.08319
3 Russian 0.02770 0.03325
4 Portuguese 0.00792 0.00747

DESCRIPTION

  • train、test共に重複レコードあり
  • ただしすべてのレコードが重複しているわけではない
print(f"[tr] len: {len(train_df)}, nunique: {train_df['DESCRIPTION'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['DESCRIPTION'].nunique()}")
[tr] len: 91333, nunique: 91307
[te] len: 91822, nunique: 91803

print(f"[tr] duplicated: {train_df['DESCRIPTION'].duplicated().sum()}")
print(f"[te] duplicated: {test_df['DESCRIPTION'].duplicated().sum()}")
[tr] duplicated: 26
[te] duplicated: 19

display(train_df.loc[train_df['DESCRIPTION'].str.startswith('Hellen is 41')])
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED LOAN_AMOUNT IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL
7196 1722176 English Hellen is 41 years old. She is married to Aurh... Hellen is 41 years old. She is married to Aurh... 225 3096942 Cattle Agriculture to buy feed and vaccines. KE Kenya Nyandarua South shared 0.10000 KES #Animals, #Woman-Owned Business, #Parent monthly field_partner
7869 1874377 English Hellen is 41 years old. She is married to Aurh... Hellen is 41 years old. She is married to Aurh... 325 3297889 Cattle Agriculture to buy animal feeds and vaccines. KE Kenya Nyandarua South shared 0.00000 KES #Woman-Owned Business, #Animals, #Repeat Borro... monthly field_partner

DESCRIPTION_TRANSLATED

  • trainに1レコードだけ欠損あり
  • train、test共に重複レコードあり
  • DESCRIPTIONと重複レコード数は異なる
display(train_df.loc[train_df['DESCRIPTION_TRANSLATED'].isnull()])
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED LOAN_AMOUNT IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL
72195 1478847 English Tursun is 62 years old and she comes from the ... NaN 800 2783623 Livestock Agriculture to buy some livestock and start a new business. TJ Tajikistan Yavan shared 0.10000 TJS #Animals, #Elderly irregular field_partner 
print(f"[tr] len: {len(train_df)}, nunique: {train_df['DESCRIPTION_TRANSLATED'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['DESCRIPTION_TRANSLATED'].nunique()}")
[tr] len: 91333, nunique: 91309
[te] len: 91822, nunique: 91808

print(f"[tr] duplicated: {train_df['DESCRIPTION_TRANSLATED'].duplicated().sum()}")
print(f"[te] duplicated: {test_df['DESCRIPTION_TRANSLATED'].duplicated().sum()}")
[tr] duplicated: 23
[te] duplicated: 14

LOAN_AMOUNT（目的変数）

  • 最小で25、最大で10000
  • 対数化するとだいたい正規分布に近い形状となる
print('[tr]')
display(train_df.describe()['LOAN_AMOUNT'])
[tr]

count   91333.00000
mean      738.14996
std      1013.75287
min        25.00000
25%       225.00000
50%       450.00000
75%       825.00000
max     10000.00000
Name: LOAN_AMOUNT, dtype: float64
plt.figure(figsize=(20, 8))
train_df['LOAN_AMOUNT'].hist();
train_df['LOAN_AMOUNT_log1p'] = np.log1p(train_df['LOAN_AMOUNT'])

plt.figure(figsize=(20, 8))
train_df['LOAN_AMOUNT_log1p'].hist();

IMAGE_ID

  • 重複なし
  • train、testでかぶっているIMAGE_IDが725レコードある
  • 同じIMAGE_IDのレコードは似ているが重複しているわけではない
print(f"[tr] len: {len(train_df)}, nunique: {train_df['IMAGE_ID'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['IMAGE_ID'].nunique()}")
[tr] len: 91333, nunique: 91333
[te] len: 91822, nunique: 91822

venn2([set(train_df['IMAGE_ID']), set(test_df['IMAGE_ID'])], set_labels=('train', 'test'));
duplicated_image_ids = list(set(train_df['IMAGE_ID']) & set(test_df['IMAGE_ID']))

display(train_df.loc[train_df['IMAGE_ID']==duplicated_image_ids[0]])
display(test_df.loc[test_df['IMAGE_ID']==duplicated_image_ids[0]])
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED LOAN_AMOUNT IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL LOAN_AMOUNT_log1p
75564 1669059 English Jennifer works hard to support five children. ... Jennifer works hard to support five children. ... 200 3028992 Personal Housing Expenses Housing to build a sanitary toilet for her family. PH Philippines Banga, Aklan shared 0.10000 PHP #Health and Sanitation, #Eco-friendly, user_fa... monthly field_partner 5.30330
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL
68693 1944221 English Jennifer works hard to support five children. ... Jennifer works hard to support five children. ... 3028992 Personal Housing Expenses Housing to build a sanitary toilet for her family. PH Philippines Banga, Aklan standard nan PHP NaN monthly field_partner 
display(train_df.loc[train_df['IMAGE_ID']==duplicated_image_ids[1]])
display(test_df.loc[test_df['IMAGE_ID']==duplicated_image_ids[1]])
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED LOAN_AMOUNT IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL LOAN_AMOUNT_log1p
19299 1806961 English Elizabeth is 47 years old, married and blessed... Elizabeth is 47 years old, married and blessed... 400 3213314 Livestock Agriculture to buy feeds, vitamins and injections for her ... PH Philippines Macate, Bambang, Nueva Vizcaya shared 0.10000 PHP #Animals monthly field_partner 5.99396
LOAN_ID ORIGINAL_LANGUAGE DESCRIPTION DESCRIPTION_TRANSLATED IMAGE_ID ACTIVITY_NAME SECTOR_NAME LOAN_USE COUNTRY_CODE COUNTRY_NAME TOWN_NAME CURRENCY_POLICY CURRENCY_EXCHANGE_COVERAGE_RATE CURRENCY TAGS REPAYMENT_INTERVAL DISTRIBUTION_MODEL
37366 1914116 English Elizabeth is known in the area as a hardworkin... Elizabeth is known in the area as a hardworkin... 3213314 Pigs Agriculture to buy feeds, vitamins, and injections for her... PH Philippines Macate, Bambang, Nueva Vizcaya shared 0.00000 PHP NaN monthly field_partner

ACTIVITY_NAME

  • ユニーク数は161
  • trainとtestで異なるACTIVITY_NAMEはない
print(f"[tr] len: {len(train_df)}, nunique: {train_df['ACTIVITY_NAME'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['ACTIVITY_NAME'].nunique()}")
[tr] len: 91333, nunique: 161
[te] len: 91822, nunique: 161

print('[tr]')
display(train_df['ACTIVITY_NAME'].value_counts())

print('-'*50)

print('[te]')
display(test_df['ACTIVITY_NAME'].value_counts())
[tr]

Farming                      11631
Personal Housing Expenses     7993
General Store                 7504
Food Production/Sales         4174
Pigs                          3790
                             ...  
Bicycle Sales                    3
Utilities                        3
Patchwork                        3
Musical Instruments              3
Balut-Making                     1
Name: ACTIVITY_NAME, Length: 161, dtype: int64
--------------------------------------------------
[te]

Farming                      13626
General Store                10318
Pigs                          6129
Personal Housing Expenses     4703
Food Production/Sales         4304
                             ...  
Balut-Making                     2
Music Discs & Tapes              2
Bookbinding                      2
Patchwork                        1
Utilities                        1
Name: ACTIVITY_NAME, Length: 161, dtype: int64
venn2([set(train_df['ACTIVITY_NAME'].unique()), set(test_df['ACTIVITY_NAME'].unique())], set_labels=('train', 'test'));

SECTOR_NAME

  • ユニーク数は161
  • trainとtestで異なるSECTOR_NAMEはない
print(f"[tr] len: {len(train_df)}, nunique: {train_df['SECTOR_NAME'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['SECTOR_NAME'].nunique()}")
[tr] len: 91333, nunique: 15
[te] len: 91822, nunique: 15

display(
    pd.DataFrame({
        'train': train_df['SECTOR_NAME'].value_counts() / len(train_df),
        'test': test_df['SECTOR_NAME'].value_counts() / len(test_df)
    }).reset_index().sort_values(['train'], ascending=False)
)
index train test
0 Agriculture 0.27334 0.31646
6 Food 0.20476 0.22642
11 Retail 0.16593 0.19337
8 Housing 0.08962 0.05245
12 Services 0.06012 0.04634
2 Clothing 0.04787 0.04273
10 Personal Use 0.04515 0.02991
4 Education 0.03868 0.03375
1 Arts 0.02165 0.01812
7 Health 0.01870 0.01563
13 Transportation 0.01631 0.00933
3 Construction 0.00910 0.00814
9 Manufacturing 0.00736 0.00632
5 Entertainment 0.00072 0.00051
14 Wholesale 0.00069 0.00052 
venn2([set(train_df['SECTOR_NAME'].unique()), set(test_df['SECTOR_NAME'].unique())], set_labels=('train', 'test'));

LOAN_USE

  • 重複あり
  • ピリオドのあり／なし、「their」と「her」だけ異なるなどあるため、前処理が必要
print(f"[tr] len: {len(train_df)}, nunique: {train_df['LOAN_USE'].nunique()}")
print(f"[te] len: {len(test_df)}, nunique: {test_df['LOAN_USE'].nunique()}")
[tr] len: 91333, nunique: 63062
[te] len: 91822, nunique: 54099

print(f"[tr] duplicated: {train_df['LOAN_USE'].duplicated().sum()}")
print(f"[te] duplicated: {test_df['LOAN_USE'].duplicated().sum()}")
[tr] duplicated: 28271
[te] duplicated: 37723

print('[tr]')
display(train_df['LOAN_USE'].value_counts())

print('-'*50)

print('[te]')
display(test_df['LOAN_USE'].value_counts())
[tr]

to build a sanitary toilet for her family                                                                 3068
to build a sanitary toilet for her family.                                                                1581
to buy a water filter to provide safe drinking water for their family.                                     895
to buy a water filter to provide safe drinking water for her family.                                       573
to buy ingredients for her food production business                                                        449
                                                                                                          ... 
to build facilities for sourcing water at home in order to improve her family's access to clean water.       1
to complete the budget needed for maintenance of his small coffee farm.                                      1
to buy corn, sugar, and milk.                                                                                1
to buy items like canned goods, personal care products, etc. to sell  in her general store.                  1
to buy more laying hens and corn to feed them.                                                               1
Name: LOAN_USE, Length: 63062, dtype: int64
--------------------------------------------------
[te]

to build a sanitary toilet for her family                                                                 1771
to buy ingredients for her food production business                                                       1057
to buy items like canned goods, personal care products, etc. to sell in her general store.                1028
to buy feed and other supplies to raise her livestock.                                                     786
to buy feed and other supplies to raise her pigs.                                                          618
                                                                                                          ... 
to pay for her university studies, get good knowledge and find a well-paid job.                              1
to acquire two bundles of used shoes wholesale, which will support her initial capital.                      1
to buy water pumps and fertilizer for his farm.                                                              1
to buy silk and poly silk thread, thimbles and dowels. She will also pay the wages of other craftsmen.       1
to purchase an additional basket of fish to sell.                                                            1
Name: LOAN_USE, Length: 54099, dtype: int64

COUNTRY_CODE

  • ユニーク数は61
  • trainとtestで異なるCOUNTRY_CODEはない
print('[tr]')
display(train_df['COUNTRY_CODE'].value_counts())

print('-'*50)

print('[te]')
display(test_df['COUNTRY_CODE'].value_counts())
[tr]

PH    21265
KE    12510
UG     5551
CO     4758
SV     4100
      ...  
PA       28
PR       16
LS       11
IL       10
AM        1
Name: COUNTRY_CODE, Length: 61, dtype: int64
--------------------------------------------------
[te]

PH    28185
KE    12444
TJ     4388
EC     3818
UG     3088
      ...  
PA       37
PR       24
LS       13
IL       11
AM        1
Name: COUNTRY_CODE, Length: 61, dtype: int64
venn2([set(train_df['COUNTRY_CODE'].unique()), set(test_df['COUNTRY_CODE'].unique())], set_labels=('train', 'test'));

COUNTRY_NAME

  • ユニーク数は61
  • trainとtestで異なるCOUNTRY_NAMEはない
  • COUNTRY_CODEとCOUNTRY_NAMEは完全に対応している
print('[tr]')
display(train_df['COUNTRY_NAME'].value_counts())

print('-'*50)

print('[te]')
display(test_df['COUNTRY_NAME'].value_counts())
[tr]

Philippines    21265
Kenya          12510
Uganda          5551
Colombia        4758
El Salvador     4100
               ...  
Panama            28
Puerto Rico       16
Lesotho           11
Israel            10
Armenia            1
Name: COUNTRY_NAME, Length: 61, dtype: int64
--------------------------------------------------
[te]

Philippines    28185
Kenya          12444
Tajikistan      4388
Ecuador         3818
Uganda          3088
               ...  
Panama            37
Puerto Rico       24
Lesotho           13
Israel            11
Armenia            1
Name: COUNTRY_NAME, Length: 61, dtype: int64
venn2([set(train_df['COUNTRY_NAME'].unique()), set(test_df['COUNTRY_NAME'].unique())], set_labels=('train', 'test'));
train_df['COUNTRY'] = train_df['COUNTRY_CODE'] + '_' + train_df['COUNTRY_NAME']
test_df['COUNTRY'] = test_df['COUNTRY_CODE'] + '_' + test_df['COUNTRY_NAME']

venn2([set(train_df['COUNTRY'].unique()), set(test_df['COUNTRY'].unique())], set_labels=('train', 'test'));

CURRENCY_POLICY

  • trainとtestでCURRENCY_POLICYの割合が異なる
display(
    pd.DataFrame({
        'train': train_df['CURRENCY_POLICY'].value_counts() / len(train_df),
        'test': test_df['CURRENCY_POLICY'].value_counts() / len(test_df)
    }).reset_index()
)
index train test
0 shared 0.89848 0.52559
1 standard 0.10152 0.47441

CURRENCY_EXCHANGE_COVERAGE_RATE

  • trainとtestでCURRENCY_EXCHANGE_COVERAGE_RATEの割合が異なる
display(
    pd.DataFrame({
        'train': train_df['CURRENCY_EXCHANGE_COVERAGE_RATE'].value_counts() / len(train_df),
        'test': test_df['CURRENCY_EXCHANGE_COVERAGE_RATE'].value_counts() / len(test_df)
    }).reset_index()
)
index train test
0 0.00000 0.04661 0.32313
1 0.10000 0.85187 0.20247

CURRENCY

  • ユニーク数は51
  • trainとtestで異なるCURRENCYはない
  • COUNTRYとCURRENCYの組み合わせでは、trainにしかない組み合わせが3組存在する
  • ただしそのレコード数は少ないため、削除しても問題ないかもしれない
print('[tr]')
display(train_df['CURRENCY'].value_counts())

print('-'*50)

print('[te]')
display(test_df['CURRENCY'].value_counts())
[tr]

PHP    21265
KES    12510
USD    11331
UGX     5551
COP     4758
TJS     3648
KHR     2768
XOF     2548
INR     2123
VND     1774
PEN     1691
PYG     1668
LRD     1642
MGA     1583
PKR     1547
WST     1352
HNL     1239
NGN     1183
KGS      980
RWF      855
JOD      784
GHS      778
GTQ      742
IDR      699
NIO      645
MZN      510
EGP      501
HTG      453
BOB      436
MXN      433
GEL      375
ALL      330
SBD      324
FJD      270
TOP      260
CRC      259
BRL      221
MDL      201
DOP      162
TRY      152
XAF      139
EUR      132
MWK      130
SLL      122
ZMW       98
NPR       62
PGK       43
THB       33
ILS       11
LSL       11
AMD        1
Name: CURRENCY, dtype: int64
--------------------------------------------------
[te]

PHP    28185
KES    12444
USD    12098
TJS     4386
XOF     3429
UGX     3088
MGA     2664
KHR     2511
COP     1965
PYG     1798
VND     1549
LRD     1462
KGS     1207
IDR     1064
NIO     1039
RWF     1010
WST      880
PEN      847
HNL      711
NGN      701
GTQ      691
JOD      674
GHS      533
SLL      495
HTG      457
EGP      454
MZN      438
ZMW      436
TOP      429
MXN      375
BOB      356
SBD      353
CRC      344
ALL      336
PKR      316
INR      308
BRL      263
FJD      259
GEL      248
MDL      224
XAF      166
MWK      127
EUR      116
DOP       93
TRY       92
PGK       63
NPR       59
THB       54
LSL       13
ILS       11
AMD        1
Name: CURRENCY, dtype: int64
venn2([set(train_df['CURRENCY'].unique()), set(test_df['CURRENCY'].unique())], set_labels=('train', 'test'));
train_df['COUNTRY_CURRENCY'] = train_df['COUNTRY_CODE'] + '_' + train_df['CURRENCY']
test_df['COUNTRY_CURRENCY'] = test_df['COUNTRY_CODE'] + '_' + test_df['CURRENCY']

venn2([set(train_df['COUNTRY_CURRENCY'].unique()), set(test_df['COUNTRY_CURRENCY'].unique())], set_labels=('train', 'test'));
# trainにのみ存在しているCOUNTRYとCURRENCYの組み合わせ
print(set(train_df['COUNTRY_CURRENCY']) - set(test_df['COUNTRY_CURRENCY'])) 
{'BO_USD', 'IN_USD', 'PS_ILS'}

print('tr:', train_df.loc[train_df['COUNTRY_CURRENCY'].str.startswith('PS'), 'COUNTRY_CURRENCY'].unique())
print('te:', test_df.loc[test_df['COUNTRY_CURRENCY'].str.startswith('PS'), 'COUNTRY_CURRENCY'].unique())
print('-'*50)
print('tr:', train_df.loc[train_df['COUNTRY_CURRENCY'].str.startswith('IN'), 'COUNTRY_CURRENCY'].unique())
print('te:', test_df.loc[test_df['COUNTRY_CURRENCY'].str.startswith('IN'), 'COUNTRY_CURRENCY'].unique())
print('-'*50)
print('tr:', train_df.loc[train_df['COUNTRY_CURRENCY'].str.startswith('BO'), 'COUNTRY_CURRENCY'].unique())
print('te:', test_df.loc[test_df['COUNTRY_CURRENCY'].str.startswith('BO'), 'COUNTRY_CURRENCY'].unique())
tr: ['PS_USD' 'PS_ILS']
te: ['PS_USD']
--------------------------------------------------
tr: ['IN_INR' 'IN_USD']
te: ['IN_INR']
--------------------------------------------------
tr: ['BO_BOB' 'BO_USD']
te: ['BO_BOB']

print(f'PS_ILS: {len(train_df.loc[train_df["COUNTRY_CURRENCY"]=="PS_ILS"])}')
print(f'IN_USD: {len(train_df.loc[train_df["COUNTRY_CURRENCY"]=="IN_USD"])}')
print(f'BO_USD: {len(train_df.loc[train_df["COUNTRY_CURRENCY"]=="BO_USD"])}')
PS_ILS: 1
IN_USD: 1
BO_USD: 17

TAGS

  • 同じタグが一つのTAGSの中に複数回書かれていることがある
  • testにのみ存在しているTAGが一つある（#US Black-Owned Business）
print('[tr]')
display(train_df['TAGS'].value_counts())

print('-'*50)

print('[te]')
display(test_df['TAGS'].value_counts())
[tr]

user_favorite                                                                                                           3961
#Woman-Owned Business                                                                                                   2814
#Repeat Borrower                                                                                                        1761
#Elderly                                                                                                                1746
#Animals                                                                                                                1586
                                                                                                                        ... 
user_favorite, #Technology, #Sustainable Ag, #Eco-friendly, #Repeat Borrower, user_favorite                                1
#Trees, user_favorite, user_favorite, #Vegan, #Sustainable Ag                                                              1
user_favorite, #Biz Durable Asset, user_favorite, #US immigrant, user_favorite, user_favorite, #Woman-Owned Business       1
user_favorite, #Parent, user_favorite, #Biz Durable Asset, #Woman-Owned Business                                           1
#Biz Durable Asset, #Animals, user_favorite, user_favorite                                                                 1
Name: TAGS, Length: 21367, dtype: int64
--------------------------------------------------
[te]

user_favorite                                                                                                                     7471
user_favorite, user_favorite                                                                                                      2100
#Parent, #Woman-Owned Business                                                                                                    1876
#Woman-Owned Business, #Parent                                                                                                    1555
#Woman-Owned Business                                                                                                             1026
                                                                                                                                  ... 
#Single Parent, user_favorite, user_favorite, #Parent, user_favorite, #Schooling, #Woman-Owned Business                              1
#Woman-Owned Business, volunteer_pick, #Repeat Borrower                                                                              1
#Elderly, user_favorite, #Woman-Owned Business, user_favorite, user_favorite, user_favorite, #Supporting Family, user_favorite       1
#Elderly, #Single                                                                                                                    1
#Single, #Supporting Family, #Eco-friendly, user_favorite, user_favorite, #Health and Sanitation, #Vegan, user_favorite              1
Name: TAGS, Length: 20112, dtype: int64
train_tags_df = train_df['TAGS'].str.get_dummies(sep=',').add_prefix('tag_')
print(f'[tr] tags nunique: {train_tags_df.shape[1]}')

test_tags_df = test_df['TAGS'].str.get_dummies(sep=',').add_prefix('tag_')
print(f'[te] tags nunique: {test_tags_df.shape[1]}')
[tr] tags nunique: 58
[te] tags nunique: 60

train_tags_df
tag_ #Animals tag_ #Biz Durable Asset tag_ #Eco-friendly tag_ #Elderly tag_ #Fabrics tag_ #Female Education tag_ #First Loan tag_ #Health and Sanitation tag_ #Job Creator tag_ #Orphan tag_ #Parent tag_ #Refugee tag_ #Repair Renew Replace tag_ #Repeat Borrower tag_ #Schooling tag_ #Single tag_ #Single Parent tag_ #Supporting Family tag_ #Sustainable Ag tag_ #Technology tag_ #Trees tag_ #US immigrant tag_ #Unique tag_ #Vegan tag_ #Widowed tag_ #Woman-Owned Business tag_ user_favorite tag_ volunteer_like tag_ volunteer_pick tag_#Animals tag_#Biz Durable Asset tag_#Eco-friendly tag_#Elderly tag_#Fabrics tag_#Female Education tag_#First Loan tag_#Health and Sanitation tag_#Job Creator tag_#Orphan tag_#Parent tag_#Refugee tag_#Repair Renew Replace tag_#Repeat Borrower tag_#Schooling tag_#Single tag_#Single Parent tag_#Supporting Family tag_#Sustainable Ag tag_#Technology tag_#Trees tag_#US immigrant tag_#Unique tag_#Vegan tag_#Widowed tag_#Woman-Owned Business tag_user_favorite tag_volunteer_like tag_volunteer_pick
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 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 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 0
2 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
3 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
4 0 0 0 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 1 0 0 0 1 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 0 0 0 0 0
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
91328 1 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 0 0 0 1 0
91329 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
91330 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 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 0
91331 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 1 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
91332 0 1 1 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 1 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

91333 rows × 58 columns

venn2([set(train_tags_df.columns), set(test_tags_df.columns)], set_labels=('train', 'test'));
# testにのみ存在しているTAGS
print(set(test_tags_df.columns) - set(train_tags_df.columns))
{'tag_ #US Black-Owned Business', 'tag_#US Black-Owned Business'}

train_tags_df.columns
Index(['tag_ #Animals', 'tag_ #Biz Durable Asset', 'tag_ #Eco-friendly',
       'tag_ #Elderly', 'tag_ #Fabrics', 'tag_ #Female Education',
       'tag_ #First Loan', 'tag_ #Health and Sanitation', 'tag_ #Job Creator',
       'tag_ #Orphan', 'tag_ #Parent', 'tag_ #Refugee',
       'tag_ #Repair Renew Replace', 'tag_ #Repeat Borrower',
       'tag_ #Schooling', 'tag_ #Single', 'tag_ #Single Parent',
       'tag_ #Supporting Family', 'tag_ #Sustainable Ag', 'tag_ #Technology',
       'tag_ #Trees', 'tag_ #US immigrant', 'tag_ #Unique', 'tag_ #Vegan',
       'tag_ #Widowed', 'tag_ #Woman-Owned Business', 'tag_ user_favorite',
       'tag_ volunteer_like', 'tag_ volunteer_pick', 'tag_#Animals',
       'tag_#Biz Durable Asset', 'tag_#Eco-friendly', 'tag_#Elderly',
       'tag_#Fabrics', 'tag_#Female Education', 'tag_#First Loan',
       'tag_#Health and Sanitation', 'tag_#Job Creator', 'tag_#Orphan',
       'tag_#Parent', 'tag_#Refugee', 'tag_#Repair Renew Replace',
       'tag_#Repeat Borrower', 'tag_#Schooling', 'tag_#Single',
       'tag_#Single Parent', 'tag_#Supporting Family', 'tag_#Sustainable Ag',
       'tag_#Technology', 'tag_#Trees', 'tag_#US immigrant', 'tag_#Unique',
       'tag_#Vegan', 'tag_#Widowed', 'tag_#Woman-Owned Business',
       'tag_user_favorite', 'tag_volunteer_like', 'tag_volunteer_pick'],
      dtype='object')

REPAYMENT_INTERVAL

  • trainとtestで割合はだいたい同じ
display(
    pd.DataFrame({
        'train': train_df['REPAYMENT_INTERVAL'].value_counts() / len(train_df),
        'test': test_df['REPAYMENT_INTERVAL'].value_counts() / len(test_df)
    }).reset_index()
)
index train test
0 monthly 0.90192 0.86572
1 bullet 0.07146 0.08050
2 irregular 0.02662 0.05378

DISTRIBUTION_MODEL

  • trainとtestで割合はだいたい同じ
display(
    pd.DataFrame({
        'train': train_df['DISTRIBUTION_MODEL'].value_counts() / len(train_df),
        'test': test_df['DISTRIBUTION_MODEL'].value_counts() / len(test_df)
    }).reset_index()
)
index train test
0 field_partner 0.99536 0.99585
1 direct 0.00464 0.00415

添付データ

  • EDA.ipynb?X-Amz-Expires=10800&X-Amz-Date=20241221T121913Z&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIP7GCBGMWPMZ42PQ
    • Icon10
    goukaisei

    神EDAありがとうございます！！！

    Aws4 request&x amz signedheaders=host&x amz signature=aa61ba011dec4a8d9defe40f25f8c5e31b4be58dde73af424b00fa244f3beccc
    yshr10ic

    ありがとうございます！ 初めてのトピック投稿なので、そう言っていただけると励みになります！

    Favicon
    new user
    コメントするには 新規登録 もしくは ログイン が必要です。