我是一名新手,并且需要使用多索引来改变我的数据框,
我的数据框如何,请注意从csv文件中读取它。
我希望它看起来如何,
请有人帮助我,我已经这么久了。我已经尝试过堆栈功能,但它根本不能正常工作。请帮帮我..
我设法做的是:
import pandas as pd
path =" C:\ Desktop \ data.csv" df = pd.read_csv(path,sep =&#39 ;;')
提前谢谢!
答案 0 :(得分:1)
您正在寻找的函数名为.stack()。 文档:https://pandas.pydata.org/pandas-docs/stable/reshaping.html#reshaping-by-stacking-and-unstacking
一个例子:
import pandas as pd
dict_ = {
"date": [2017,2016,2015,2014],
"total members": [150,140,100,150],
"lions": [20,10,5,50]
}
df = pd.DataFrame.from_dict(dict_)
table = df.set_index("date").stack() # sets date as index
table.to_frame(name='Amount').to_excel('output.xlsx') # names the column: Amount and outputs to an xlsx file
打印:
date
2017 lions 20
total members 150
2016 lions 10
total members 140
2015 lions 5
total members 100
2014 lions 50
total members 150
dtype: int64
对原始帖子的评论
将来,您可以使用以下方式轻松共享数据框:
df.head().to_dict() # .head() return first 5 rows
此示例输出:
{'date': {0: 2017, 1: 2016, 2: 2015, 3: 2014},
'lions': {0: 20, 1: 10, 2: 5, 3: 50},
'total members': {0: 150, 1: 140, 2: 100, 3: 150}}
答案 1 :(得分:1)
让我们使用set_index
和import pandas as pd
from io import StringIO
csv_file = StringIO("""date,2017,2016,2015,2014
total members, 150, 140, 100, 150
lions, 20, 10, 5, 25
demokrats, 60, 50, 15, 25
liberals, 20, 5, 40, 25
partizans, 50, 75, 40, 25""")
df = pd.read_csv(csv_file)
df_out= (df.rename(columns={'date':'Parties'})
.melt(id_vars='Parties',var_name='Date',value_name='Amount')
.set_index(['Date','Parties']))
print(df_out)
:
Amount
Date Parties
2017 total members 150
lions 20
demokrats 60
liberals 20
partizans 50
2016 total members 140
lions 10
demokrats 50
liberals 5
partizans 75
2015 total members 100
lions 5
demokrats 15
liberals 40
partizans 40
2014 total members 150
lions 25
demokrats 25
liberals 25
partizans 25
输出:
safariOptions.addExtensions("..");