如何在Pandas中将DataFrame列表列的元素转换为字符串类型

Hey there! Let's break down the issue with your current code and fix it to get exactly the result you want.

Why Your Current Code Isn't Working #

The line df3['label'] = [str(i) for x in df3['label'] for i in x] is flattening all elements from your label lists into a single 1D list instead of preserving each row's list structure. For example, with your sample data, this would create a long list like ['0','0','1','1','0','0','1','1','316.0',...]—which has way more elements than your 266-row DataFrame. This mismatch is why it fails to produce the output you need.

Simple Fixes to Preserve List Structure #

You have two straightforward, effective ways to convert each element in the label lists to strings while keeping the per-row list intact:

Option 1: Use pandas.Series.apply() with a Lambda

This is a clean, pandas-native approach:

df3['label'] = df3['label'].apply(lambda x: [str(i) for i in x])

The apply() method runs the lambda function on every list in the label column. The lambda takes each list x, converts every element i to a string, and returns the modified list for that row.

Option 2: Nested List Comprehension

If you prefer list comprehensions, adjust the structure to keep each row's list intact:

df3['label'] = [[str(i) for i in x] for x in df3['label']]

Here, the outer comprehension loops through each list in df3['label'], and the inner comprehension converts each element in that sublist to a string—creating a new, string-filled list for every row.

Test with Your Sample Data #

Let's confirm this works with your example:

import pandas as pd

# Sample data matching your structure
data = {
    'Id': ['4cfab', '924cf', '916b8', '0f549', 'db696'],
    'label': [[0,0,1,1], [0,0,1,1], [316.0,318.0], [0,0,1,1], [345.0,39.0,9]]
}
df3 = pd.DataFrame(data)

# Apply the fix (either option works!)
df3['label'] = df3['label'].apply(lambda x: [str(i) for i in x])

print(df3)

This outputs exactly the result you're targeting:

Id                  label
0  4cfab        ['0', '0', '1', '1']
1  924cf        ['0', '0', '1', '1']
2  916b8          ['316.0', '318.0']
3  0f549        ['0', '0', '1', '1']
4  db696  ['345.0', '39.0', '9']

内容的提问来源于stack exchange，提问作者pasan