如何将几个单词列表转换为pandas数据帧？

Question

我有文件.txt包含这样的单词列表:

5.91686268506 exclusively, catering, provides, arms, georgia, formal, purchase, choose
5.91560417296 hugh, senlis
5.91527936181 italians
5.91470429433 soil, cultivation, fertile
5.91468087491 increases, moderation
....
5.91440227412 farmers, descendants

我想将这些数据转换为pandas表,我希望将其显示到html/bootstrap模板中,如下所示(*):

COL_A         COL_B
5.91686268506 exclusively, catering, provides, arms, georgia, formal, purchase, choose
5.91560417296 hugh, senlis
5.91527936181 italians
5.91470429433 soil, cultivation, fertile
5.91468087491 increases, moderation
....
5.91440227412 farmers, descendants

所以我用熊猫尝试了以下内容:

import pandas as pd
df = pd.read_csv('file.csv', 
                 sep = ' ', names=['Col_A', 'Col_B'])
df.head(20)

但是,我的表没有上述所需的结构:

                                                                                                                                COL_A   COL_B
6.281426    engaged,    chance,     makes,  meeting,    nations,    things,     believe,    tries,  believing,  knocked,    admits,     awkward
6.277438    sweden  NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN
6.271190    artificial,     ammonium    NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN
6.259790    boats,  prefix  NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN     NaN
6.230612    targets,    tactical,   wing,   missile,    squadrons   NaN     NaN     NaN     NaN     NaN     NaN     NaN

有关如何以(*)表格格式获取数据的任何想法？

Answer 1

因为单词之间有空格,如果将空格指定为分隔符,它自然会将它们分开.为了得到你需要的东西,你可以尝试将其设置sep为正则表达式(?<!,),?<!是语法背后的负面看法,这意味着只有在空格前面没有逗号时它才会分开,它应该适用于你的情况:

pd.read_csv("~/test.csv", sep = "(?<!,) ", names=['weight', 'topics'])

#     weight    topics
#0  5.916863    exclusively, catering, provides, arms, georgia...
#1  5.915604    hugh, senlis
#2  5.915279    italians
#3  5.914704    soil, cultivation, fertile
#4  5.914681    increases, moderation
#5  5.914402    farmers, descendants