索引和搜索熊猫数据框内的列表列表

Question

Let say we have a list of lists like this:

lst = [[[[1,2]]],[[[2,3]]],[[[3,4]]]]

and then we make a panda data frame here:

import pandas as pd
import numpy as np
df = pd.DataFrame(lst, columns = ['col'])
print df

and returns

        col
0  [[1, 2]]
1  [[2, 3]]
2  [[3, 4]]

what I want is replacing col with green when the interior list includes 1 and red if not. I tried this:

df['col'] = np.where(1 in df['col'], 'green', 'red')
print df

but does not return correct answer!!!

     col
0  green
1  green
2  green

I think the problem is that each element of data frame is a list of lists.

Answer 1

You can use apply method; Also you have a list of list of list, you can either use str[0] to extract the first element (of list type) out or you can extract it from x in apply :

df['col'].str[0].apply(lambda x: 'green' if 1 in x else 'red')

#0    green
#1      red
#2      red
#Name: col, dtype: object

Another option:

df['col'].apply(lambda x: 'green' if 1 in x[0] else 'red')

#0    green
#1      red
#2      red
#Name: col, dtype: object