Double For 循环（通过 DataFrame 和 List）答案

【问题标题】：Double ForLooping (over DataFrame & List)Double For 循环（通过 DataFrame 和 List）
【发布时间】：2021-01-21 22:01:03
【问题描述】：

我有这个 test2 数据框：

   manufacturer  condition   fuel        drive  cylinders       description
0   ford        excellent    gas          rwd    NaN            ford in excellent condition. 4 cylinders
1   cadillac    NaN          NaN          NaN    4 cylinders    4 cylinders. Half-new cadillac. Diesel.
2   NaN         new          diesel       fwd    12 cylinders   Ford, diesel, new condition.
3   NaN         NaN          electric     NaN    10 cylinders   Ferrari, excellent condition. 4wd
4   ferrari     NaN          NaN          4wd    NaN            New ferrari. Electric with 12 cylinders.

我想遍历数据框并使用“描述”列的信息填充每列的 NaN 值。为此，我这样做了：

import re

manufacturer = '(ford | cadillac | ferrari)'
condition = '(excellent, good, fair, like new, salvage, new)'
fuel = '(gas, hybrid, diesel, electric)'
drive = '(\S*wd)'
cylinders = '(\d+\s+cylinders?)'

test2['manufacturer'] = test2['manufacturer'].fillna(
    test2['description'].str.extract(manufacturer, flags=re.IGNORECASE, expand=False)).str.lower()
test2['condition'] = test2['condition'].fillna(
    test2['description'].str.extract(condition, flags=re.IGNORECASE, expand=False)).str.lower()
test2['fuel'] = test2['fuel'].fillna(
    test2['description'].str.extract(fuel, flags=re.IGNORECASE, expand=False)).str.lower()
test2['drive'] = test2['drive'].fillna(
    test2['description'].str.extract(drive, flags=re.IGNORECASE, expand=False)).str.lower()
test2['cylinders'] = test2['cylinders'].fillna(
    test2['description'].str.extract(cylinders, flags=re.IGNORECASE, expand=False)).str.lower()

test2

但它看起来不太好，所以我尝试做一个 for 循环来简化编程：

columns = [manufacturer, condition, fuel, drive, cylinders]

for i in test2:
   for column in columns:
      test2[i] = test2[i].fillna(
        test2['description'].str.extract(column, flags=re.IGNORECASE, expand=False)).str.lower()

无论我如何尝试，它都会不断给我错误。它在 test2 中的“i”上循环正常，但是当它开始在列表“columns”上循环时，循环会出错......

知道我该如何解决这个问题吗？谢谢！

【问题讨论】：

该列表应称为columns，而不是column
啊对不起，它叫列...

标签： python dataframe for-loop nan

【解决方案1】：

每个元素循环多次。你应该每个元素只循环一次。使用zip函数`合并键和列表。

试试这个代码：

keys =    ['manufacturer', 'condition', 'fuel', 'drive', 'cylinders']
columns = [ manufacturer,   condition,   fuel,   drive,   cylinders]

for i,column in zip(keys,columns):
   test2[i] = test2[i].fillna(
      test2['description'].str.extract(column, flags=re.IGNORECASE, expand=False)).str.lower()

【讨论】：