Question

我遇到了Twitter用户(Kaggle Source)的数据集，但我发现数据集的格式相当奇怪。它包含一个包含列标题的行，然后包含基本上是JSON数组的行。数据集也很大，这使得将整个文件转换为JSON对象非常困难。

将这些数据加载到Python中的好方法是什么，最好是Pandas Dataframe？

数据示例

id,screenName,tags,avatar,followersCount,friendsCount,lang,lastSeen,tweetId,friends
"1969527638","LlngoMakeEmCum_",[ "#nationaldogday" ],"http://pbs.twimg.com/profile_images/534286217882652672/FNmiQYVO_normal.jpeg",319,112,"en",1472271687519,"769310701580083200",[ "1969574754", "1969295556", "1969284056", "1969612214", "1970067476", "1969797386", "1969430539", "1969840064", "1969698176", "1970005154", "283011644", "1969901029", "1969563175", "1969302314", "1969978662", "1969457936", "1969667533", "1969547821", "1969943478", "1969668032", "283006529", "1969809440", "1969601096", "1969298856", "1969331652", "1969385498", "1969674368", "1969565263", "1970144676", "1969745390", "1969947438", "1969734134", "1969801326", "1969324008", "1969259820", "1969535827", "1970072989", "1969771688", "1969437804", "1969507394", "1969509972", "1969751588", "283012808", "1969302888", "1970224440", "1969603532", "283011244", "1969501046", "1969887518", "1970153138", "1970267527", "1969941955", "1969421654", "1970013110", "1969544905", "1969839590", "1969876500", "1969674625", "1969337952", "1970046536", "1970090934", "1969419133", "1969517215", "1969787869", "1969298065", "1970149771", "1969422638", "1969504268", "1970025554", "1969776001", "1970138611", "1969316186", "1969547558", "1969689272", "283009727", "283015491", "1969526874", "1969662210", "1969536164", "1969320008", "1969893793", "1970158393", "1969365936", "1970194418", "1969942094", "1969631580", "1969704756", "1969920092", "1969712882", "1969791680", "1969408164", "1969754851", "1970205480", "1969840267", "1969443211", "1969706762", "1969692698", "1969751576", "1969486796", "1969286630", "1969686674", "1969833492", "1969294814", "1969472719", "1969685018", "283008559", "283011243", "1969680078", "1969545697", "1969646412", "1969442725", "1969692529" ]
"51878493","_notmichelle",[ "#nationaldogday" ],"http://pbs.twimg.com/profile_images/761977602173046786/4_utEHsD_normal.jpg",275,115,"en",1472270622663,"769309490038439936",[ "60789485", "2420931980", "2899776756", "127410795", "38747286", "1345516880", "236076395", "1242946609", "2567887488", "280777286", "2912446303", "1149916171", "3192577639", "239569380", "229974168", "389097282", "266336410", "1850301204", "2364414805", "812302213", "2318240348", "158634793", "542282350", "569664772", "766573472", "703551325", "168564432", "261054460", "402980453", "562547390", "539630318", "165167145", "22216387", "427568285", "61033129", "213519434", "373092437", "170762012", "273601960", "322108757", "1681816280", "357843027", "737471496", "406541143", "1084122632", "633477616", "537821327", "793079732", "2386380799", "479015607", "783354019", "365171478", "625002575", "2326207404", "1653286842", "1676964216", "2296617326", "1583692190", "1315393903", "377660026", "2235123476", "792779641", "351222527", "444993309", "588396446", "377629159", "469383424", "1726612471", "415230430", "942443390", "360924168", "318593248", "565022085", "319679735", "632508305", "377638254", "1392782078", "584483723", "377703135", "180463340", "564978577", "502517645", "1056960042", "285097108", "410245879", "159121042", "570399371", "502348447", "960927356", "377196638", "478142245", "335043809", "73546116", "11348282", "901302409", "53255593", "515983155", "391774800", "62351523", "724792351", "346296289", "152520627", "559053427", "508019115", "349996133", "378859519", "65120103", "190070557", "339868374", "417355200", "256729771", "16171898", "45266183", "16143507", "165258639" ]

Answer 1

我们可以从这样的事情开始：（可能需要重新考虑使用|。我们可以选择更具异国情调的内容，例如╡

import pandas as pd
import io
import json

data = '''\
id,screenName,tags,avatar,followersCount,friendsCount,lang,lastSeen,tweetId,friends
"1969527638","LlngoMakeEmCum_",[ "#nationaldogday" ],"http://pbs.twimg.com/profile_images/534286217882652672/FNmiQYVO_normal.jpeg",319,112,"en",1472271687519,"769310701580083200",[ "1969574754", "1969295556", "1969284056", "1969612214", "1970067476", "1969797386", "1969430539", "1969840064", "1969698176", "1970005154", "283011644", "1969901029", "1969563175", "1969302314", "1969978662", "1969457936", "1969667533", "1969547821", "1969943478", "1969668032", "283006529", "1969809440", "1969601096", "1969298856", "1969331652", "1969385498", "1969674368", "1969565263", "1970144676", "1969745390", "1969947438", "1969734134", "1969801326", "1969324008", "1969259820", "1969535827", "1970072989", "1969771688", "1969437804", "1969507394", "1969509972", "1969751588", "283012808", "1969302888", "1970224440", "1969603532", "283011244", "1969501046", "1969887518", "1970153138", "1970267527", "1969941955", "1969421654", "1970013110", "1969544905", "1969839590", "1969876500", "1969674625", "1969337952", "1970046536", "1970090934", "1969419133", "1969517215", "1969787869", "1969298065", "1970149771", "1969422638", "1969504268", "1970025554", "1969776001", "1970138611", "1969316186", "1969547558", "1969689272", "283009727", "283015491", "1969526874", "1969662210", "1969536164", "1969320008", "1969893793", "1970158393", "1969365936", "1970194418", "1969942094", "1969631580", "1969704756", "1969920092", "1969712882", "1969791680", "1969408164", "1969754851", "1970205480", "1969840267", "1969443211", "1969706762", "1969692698", "1969751576", "1969486796", "1969286630", "1969686674", "1969833492", "1969294814", "1969472719", "1969685018", "283008559", "283011243", "1969680078", "1969545697", "1969646412", "1969442725", "1969692529" ]
"51878493","_notmichelle",[ "#nationaldogday" ],"http://pbs.twimg.com/profile_images/761977602173046786/4_utEHsD_normal.jpg",275,115,"en",1472270622663,"769309490038439936",[ "60789485", "2420931980", "2899776756", "127410795", "38747286", "1345516880", "236076395", "1242946609", "2567887488", "280777286", "2912446303", "1149916171", "3192577639", "239569380", "229974168", "389097282", "266336410", "1850301204", "2364414805", "812302213", "2318240348", "158634793", "542282350", "569664772", "766573472", "703551325", "168564432", "261054460", "402980453", "562547390", "539630318", "165167145", "22216387", "427568285", "61033129", "213519434", "373092437", "170762012", "273601960", "322108757", "1681816280", "357843027", "737471496", "406541143", "1084122632", "633477616", "537821327", "793079732", "2386380799", "479015607", "783354019", "365171478", "625002575", "2326207404", "1653286842", "1676964216", "2296617326", "1583692190", "1315393903", "377660026", "2235123476", "792779641", "351222527", "444993309", "588396446", "377629159", "469383424", "1726612471", "415230430", "942443390", "360924168", "318593248", "565022085", "319679735", "632508305", "377638254", "1392782078", "584483723", "377703135", "180463340", "564978577", "502517645", "1056960042", "285097108", "410245879", "159121042", "570399371", "502348447", "960927356", "377196638", "478142245", "335043809", "73546116", "11348282", "901302409", "53255593", "515983155", "391774800", "62351523", "724792351", "346296289", "152520627", "559053427", "508019115", "349996133", "378859519", "65120103", "190070557", "339868374", "417355200", "256729771", "16171898", "45266183", "16143507", "165258639" ]'''

# Create new separator (|) after 9th comma (',')
data = '\n'.join(['|'.join(row.split(',',9)) for row in data.split('\n')])

# REPLACE WITH THIS FOR REAL FILE
#with open('path/to/file') as f:
    #data = '\n'.join(['|'.join(row.split(',',9)) for row in f.read().split('\n')])

# Read dataframe
df = pd.read_csv(io.StringIO(data), sep='|')

# Convert strings to objects with json module:
df['friends'] = df['friends'].apply(lambda x: json.loads(x))
df['tags'] = df['tags'].apply(lambda x: json.loads(x))

更安全的方法：

import pandas as pd
import io
import json

with open('path/to/file') as f:
    columns, *rows = [row.split(',',9) for row in f.read().split('\n')]

df = pd.DataFrame(rows, columns=columns)

# Convert strings to objects with json module:
df['friends'] = df['friends'].apply(lambda x: json.loads(x))
df['tags'] = df['tags'].apply(lambda x: json.loads(x))

如何使用嵌套数组加载CSV

1 个答案: