在dataframe python中组合不同行中的列

I am trying to join columns in different rows in a dataframe.

我正在尝试在数据框中的不同行中连接列。

import pandas as pd

tdf =  {'ph1': [1, 2], 'ph2': [3, 4], 'ph3': [5,6], 'ph4': [nan,nan]}

df = pd.DataFrame(data=tdf)

df

Output:

   ph1  ph2  ph3  ph4

0    1    3    5  nan

1    2    4    6  nan

I combined ph1, ph2, ph3, ph4 with below code:

我将ph1,ph2,ph3,ph4与以下代码结合起来:

for idx, row in df.iterrows():

        df = df[[ph1, ph2, ph3, ph4]]

        df["ConcatedPhoneNumbers"] = df.loc[0:].apply(lambda x: ', '.join(x), axis=1)

I got

df["ConcatPhoneNumbers"]

ConcatPhoneNumbers

1,3,5,,

2,4,6,,

Now I need to combine these columns using pandas with appropriate function. My result should be 1,3,5,2,4,6

现在我需要使用具有适当功能的pandas来组合这些列。我的结果应该是1,3,5,2,4,6

Also need to remove these extra commas.

还需要删除这些额外的逗号。

I am new Python learner.I did some research and reached till here. Please help me to get the exact approach.

我是新的Python学习者。我做了一些研究,直到这里。请帮我准确一点。

1 个解决方案

#1

It seems you need stack for remove NaNs, then convert to int, str and list and last join:

看来你需要堆栈来删除NaNs,然后转换为int,str和list以及last join:

tdf =  {'ph1': [1, 2], 'ph2': [3, 4], 'ph3': [5,6], 'ph4': [np.nan,np.nan]}

df = pd.DataFrame(data=tdf)

cols = ['ph1', 'ph2', 'ph3', 'ph4']
s = ','.join(df[cols].stack().astype(int).astype(str).values.tolist())
print (s)
1,3,5,2,4,6