使用 NaN 条件迭代数据帧

Question

我有一个 table，我想从 table A 中提取数据并将其整理为 table B:

Table答：

Day	City A	City B	City C
Mon	NaN	Mike	NaN
Tue	NaN	NaN	Joe
Wed	Jack	Charlie	NaN

Table乙：

Day	Name	City
Mon	Mike	City B
Tue	Joe	City C
Wed	Jack	City A
Wed	Charlie	City B

我从 excel sheet 中提取了这些信息，并且正在使用 python 来完成这项任务。我的想法是我需要将数据作为数据框绘制，遍历行以查找不包含 NaN 的条目并将它们的位置和关联数据存储在新数据框中。

不幸的是，我在设置忽略 NaN 条目的条件时卡住了，我正在尝试逐步测试它并取得了进展：

    import pandas as pd
    df = pd.read_excel('./csvtasks/rosta.xlsx',sheet_name='Sheet2')
    
    #open new excel to write to with new variable df2
    #determine whether null or not
    dg=df.notnull()
    #loop over rows
    for i,j in dg.iteritems():
        if dg.bool==FALSE:
            print('skipped something') #i want this to skip but using this print to see if it's actually skipped anything
        else:
            print (i,j)
            #this will be replace by some command that uses the df.iloc[something] and writes to df2, printing for now so i can see what it does
    #loop to end
    #close file

所有这些所做的就是把整个数据框作为一个 bool 像这样给我：

Day	City A	City B	City C
True	False	True	False
True	False	False	True
True	True	True	False

Answer 1

试试 stack

s = df.set_index('Day').stack().reset_index()
s.columns = ['Day','City','Name']
s
Out[43]: 
   Day    City     Name
0  Mon  City B     Mike
1  Tue  City C      Joe
2  Wed  City A     Jack
3  Wed  City B  Charlie

Answer 2

您可以尝试 melt 然后 dropna

out = (df.melt(id_vars='Day', var_name='Name', value_name='City')
       .dropna())

print(out)

   Day    Name     City
2  Wed  City A     Jack
3  Mon  City B     Mike
5  Wed  City B  Charlie
7  Tue  City C      Joe

使用 NaN 条件迭代数据帧

Iterate over dataframe with NaN condition

python

pandas

dataframe

conditional-statements

nan