4

我的 CSV 包含有关个人及其死亡日期和死亡区域的信息。我也在使用 geopy 来定位这个区域。我认为这部分正在工作，但我真的不想拥有个人，但我想要每天和每个区域的每个事件的总人数。因此，它给了我一个 CSV，其中包含有关死亡人数、地理位置、日期以及它发生的地区/省份的信息。这是我的 csv 的片段：

rowid,Name,Status,Sex,Province,Area,Date_of_Death,Cause_of_Death
1,Hasan Ali Saad al-Hareeri ,Civilian,Adult - Male,Daraa,Sawara,2015-12-10,Warplane shelling
2,Riham Adnan Swais ,Civilian,Adult - Female,Homs,Telbeiseh,2015-12-10,Shelling
3,Ammar Ahmad al-Njoam ,Civilian,Adult - Male,Aleppo,Jarablus,2015-12-10,Explosion
4,Yaseen Slaiman al-Salkhadi ,Civilian,Adult - Male,Daraa,Jassim,2015-12-10,Warplane shelling
5,Ayman Noman Qanatra ,Civilian,Adult - Male,Daraa,Jassim,2015-12-10,Warplane shelling
6,Abdul Kareem Mohammad al-Tamki ,Civilian,Adult - Male,Daraa,Sheikh Miskeen,2015-12-10,Warplane shelling
7,Ahmad Mohammad al-Tamki ,Civilian,Child - Male,Daraa,Sheikh Miskeen,2015-12-10,Warplane shelling
8,Hamze Mohammad al-Tamki ,Civilian,Adult - Male,Daraa,Sheikh Miskeen,2015-12-10,Warplane shelling
12,Ahmad Ibraheem al-Sayed Omar ,Civilian,Adult - Male,Aleppo,,2015-12-10,Warplane shelling
13,Unidentified ,Civilian,Child - Male,Aleppo,Kafrnaya,2015-12-10,Warplane shelling
14,Unidentified 1 ,Civilian,Adult - Female,Aleppo,Bazzaa,2015-12-10,Warplane shelling
15,Unidentified 2 ,Civilian,Child - Male,Aleppo,Bazzaa,2015-12-10,Warplane shelling
16,Unidentified 3 ,Civilian,Child - Male,Aleppo,Bazzaa,2015-12-10,Warplane shelling
17,Mohammad Ahmad al-Farhoud ,Civilian,Adult - Male,Idlib,Telminis,2015-12-10,Shooting
23,wife of Faisal al-Ahmad al-Khaleefe ,Civilian,Adult - Female,Deir Ezzor,Qourieh,2015-12-10,Warplane shelling
24,Unidentified ,Civilian,Adult - Male,Deir Ezzor,Takaya,2015-12-10,Kidnapping - Execution
25,Unidentified 1 ,Civilian,Adult - Male,Deir Ezzor,Tabya Jazera village,2015-12-10,Kidnapping - Execution
26,Unidentified 2 ,Civilian,Adult - Male,Deir Ezzor,Tabya Jazera village,2015-12-10,Kidnapping - Execution
27,Unidentified 3 ,Civilian,Adult - Male,Deir Ezzor,Tabya Jazera village,2015-12-10,Kidnapping - Execution
28,Unidentified 4 ,Civilian,Adult - Male,Deir Ezzor,Tabya Jazera village,2015-12-10,Kidnapping - Execution
29,Unidentified 5 ,Civilian,Adult - Male,Deir Ezzor,Tabya Jazera village,2015-12-10,Kidnapping - Execution
32,Amoun Murad ,Civilian,Adult - Female,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
33,Kafa Mezweq ,Civilian,Adult - Female,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
34,Hala Mohammad ,Civilian,Adult - Female,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
35,Ahmad Mezweq ,Civilian,Child - Male,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
36,Ahlam Mezweq ,Civilian,Child - Female,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
37,Adnan Hussain Haj Kheder ,Civilian,Child - Male,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
38,Wife of Hussain Haj Kheder ,Civilian,Adult - Female,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
39,daughter of Hussain Haj Kheder 1 ,Civilian,Child - Female,Idlib,Jisr Shagour: Mraand,2015-12-09,Warplane shelling
40,Muhammad Muhammad al-Aoutani ,Civilian,Adult - Male,Damascus Suburbs,Hamouria,2015-12-09,Warplane shelling
41,Ibrahim al-Shami ,Civilian,Adult - Male,Damascus Suburbs,Hamouria,2015-12-09,Warplane shelling

这是我的代码

import csv
from geopy.geocoders import GoogleV3

geolocator = GoogleV3() #here some parameters are needed
lookups = {}
valueCounter = 0
init= True


with open('151213_Martyrs_filtered_civilian_2015_with_country.csv', 'rb') as csvinput:
    with open('151213_output_martyrs_final.csv', 'w') as csvoutput:
        output_fieldnames = ['Value','Country','Province','Area','Date_of_Death','Cause_of_Death','Latitude','Longitude']
        writer = csv.DictWriter(csvoutput, delimiter=',', fieldnames=output_fieldnames)
        reader = csv.DictReader(csvinput)
        for row in reader:
            if init == True :
                prevProvince = row['Province']
                prevArea = row['Area']
                prevDate = row['Date_of_Death']
                prevCause = row['Cause_of_Death']
                init = False
            if prevArea == row['Area'] and prevCause == row['Cause_of_Death'] and prevDate['Date_of_Death']:
                valueCounter +=1

                if row['Province'].split(' ', 1) == 'Damascus':
                    rowProv = "Damascus"
                else:
                    rowProv = row['Province']
                #here you have to replace the dict item by your csv column names
                query = ','.join(str(x) for x in (rowProv, row['Area']))

                try:
                    address, (latitude, longitude) = lookups[query]
                except KeyError: # new lookup
                    try:
                        query = ','.join(str(x) for x in (rowProv, rowProv))
                        address, (latitude, longitude) = lookups[query]
                        # print('lookup')
                    except KeyError: # search
                        try:
                            query = ','.join(str(x) for x in (rowProv, row['Area']))
                            address, (latitude, longitude) = geolocator.geocode(query)
                        except: # error
                            query = ','.join(str(x) for x in (rowProv, rowProv))
                            address, (latitude, longitude) = geolocator.geocode(query)
                        lookups[query] = address, (latitude, longitude)
                        print(query)

            else: #write down and save previous Row data
                output_row = {}
                output_row['Value'] = valueCounter
                output_row['Province'] = prevProvince
                output_row['Area'] = prevArea
                output_row['Date_of_Death'] = prevDate
                output_row['Cause_of_Death'] = prevCause
                output_row['Latitude'] = latitude
                output_row['Longitude'] = longitude
                writer.writerow(output_row)

                prevProvince = row['Province']
                prevArea = row['Area']
                prevDate = row['Date_of_Death']
                prevCause = row['Cause_of_Death']
                valueCounter = 1


                if row['Province'].split(' ', 1) == 'Damascus':
                    rowProv = "Damascus"
                else:
                    rowProv = row['Province']
                #here you have to replace the dict item by your csv column names
                query = ','.join(str(x) for x in (rowProv, row['Area']))

                try:
                    address, (latitude, longitude) = lookups[query]
                except KeyError: # new lookup
                    try:
                        query = ','.join(str(x) for x in (rowProv, rowProv))
                        address, (latitude, longitude) = lookups[query]
                        # print('lookup')
                    except KeyError: # search
                        try:
                            query = ','.join(str(x) for x in (rowProv, row['Area']))
                            address, (latitude, longitude) = geolocator.geocode(query)
                        except: # error
                            query = ','.join(str(x) for x in (rowProv, rowProv))
                            address, (latitude, longitude) = geolocator.geocode(query)
                        lookups[query] = address, (latitude, longitude)
                        print(query)

我没有看到什么，因为它不起作用？

终端给了我这个消息

    Test_stackoverflow_cached Ian$ python geopy_script_bodycount.py
Traceback (most recent call last):
  File "geopy_script_bodycount.py", line 22, in <module>
    if prevArea == row['Area'] and prevCause == row['Cause_of_Death'] and prevDate['Date_of_Death']:
TypeError: string indices must be integers, not str
4

2 回答 2

3

我不确定您在该行中要做什么，但问题出在跟踪中提到的那个：

if prevArea == row['Area'] and prevCause == row['Cause_of_Death'] and prevDate['Date_of_Death']:

这prevDate是一个字符串，而不是像row变量这样的字典。您可能想比较它，就像在其他子句中一样。

于 2015-12-20T19:04:21.920 回答
3

pandas使用python 数据咀嚼和分析库，做这种事情变得容易得多。

这是一个很好的介绍：http: //youtube.com/watch?v= otCriSKVV_8

这将使您开始解决问题：

import pandas as pd

data = pd.read_csv(
    'data.csv',
    index_col='rowid',
    parse_dates=['Date_of_Death'],
)

# How many died at each day?
print('Deaths by date')
print(data.groupby('Date_of_Death').size())

# How many per area:
print('\nDeaths by Area')
print(data.groupby('Area').size())

# Or both:
print('\nPer date and Area')
print(data.groupby(['Date_of_Death', 'Area']).size())
于 2015-12-20T23:28:47.993 回答