英文:
How can we melt a dataframe and list words under columns?
问题
我有一个看起来像这样的数据框。
import pandas as pddata = {'clean_words':['good','evening','how','are','you','how','can','i','help'],'start_time':[1900,2100,2500,2750,2900,1500,1650,1770,1800],'end_time':[2100,2500,2750,2900,3000,1650,1770,1800,1950],'transaction':[1,1,1,1,1,2,2,2,2]}df = pd.DataFrame(data)df
如果我尝试基本的melt操作，如下所示...
df_melted = df.pivot_table(index='clean_words', columns='transaction')df_melted.tail()
我得到这个...
我真正想要的是将交易号作为列，然后按单词列出。因此，如果transaction1是列，这些单词将在该列下列出：
'good','evening','how','are','you'
在transaction2下，这些单词将在该列下列出：
'how','can','i','help'
我该如何做呢？这里的start_time和end_time有点多余。
英文:
I have a dataframe that looks like this.
import pandas as pddata = {'clean_words':['good','evening','how','are','you','how','can','i','help'],'start_time':[1900,2100,2500,2750,2900,1500,1650,1770,1800],'end_time':[2100,2500,2750,2900,3000,1650,1770,1800,1950],'transaction':[1,1,1,1,1,2,2,2,2]}df = pd.DataFrame(data)df
If I try a basic melt, like so...
df_melted = df.pivot_table(index='clean_words', columns='transaction')df_melted.tail()
I get this...
What I really want is the transaction number as columns and the words listed down. So, if transaction1 was the column, these words would be listed in rows, under that column:
`'good','evening','how','are','you'`
Under transaction2, these words would be listed in rows, under that column:
'how','can','i','help'
How can I do that? The start_time and end_time are kind of superfluous here.
答案1
得分: 1
这是您想要的格式吗？
>>> pd.DataFrame({'1': ['good', 'evening', 'how', 'are', 'you'], '2': ['how', 'can', 'I', 'help', None]})1 20 good how1 evening can2 how I3 are help4 you None
我以后可以将您提供的内容翻译成中文。
英文:
Is this the format you want?
>>> pd.DataFrame({'1': ['good', 'evening', 'how', 'are', 'you'], '2': ['how', 'can', 'I', 'help', None]})1 20 good how1 evening can2 how I3 are help4 you None
I haven't done that before but you could pivot your data and collect a list of words under each transaction column.
>>> df.pivot_table(columns='transaction', values='clean_words', aggfunc=list)transaction 1 2clean_words [good, evening, how, are, you] [how, can, i, help]
Or group by transaction and collect a list of words.
>>> df.groupby('transaction', as_index=False).agg(clean_words=pd.NamedAgg(column='clean_words', aggfunc=list))transaction clean_words0 1 [good, evening, how, are, you]1 2 [how, can, i, help]
答案2
得分: 1
import pandas as pdimport numpy as npdata = {'clean_words': ['good', 'evening', 'how', 'are', 'you', 'how', 'can', 'i', 'help'],'start_time': [1900, 2100, 2500, 2750, 2900, 1500, 1650, 1770, 1800],'end_time': [2100, 2500, 2750, 2900, 3000, 1650, 1770, 1800, 1950],'transaction': [1, 1, 1, 1, 1, 2, 2, 2, 2]}df = pd.DataFrame(data)df_melted = df.groupby('transaction')['clean_words'].apply(np.array).reset_index()print(df_melted)
英文:
import pandas as pdimport numpy as npdata = {'clean_words':['good','evening','how','are','you','how','can','i','help'],'start_time':[1900,2100,2500,2750,2900,1500,1650,1770,1800],'end_time':[2100,2500,2750,2900,3000,1650,1770,1800,1950],'transaction':[1,1,1,1,1,2,2,2,2]}df = pd.DataFrame(data)df_melted = df.groupby('transaction')['clean_words'].apply(np.array).reset_index()print(df_melted)transaction clean_words0 1 [good, evening, how, are, you]1 2 [how, can, i, help]
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论