Pandas中聚合DataFrame列表字段报错：无数值类型可聚合

Why This Happens #

The error you're seeing makes total sense: your C column contains lists instead of numeric values (like integers or floats). Pandas' built-in aggregation functions like mean and std only work with numeric data types—they don't know how to handle lists directly, hence the DataError.

Solutions #

Here are two practical approaches to fix this, depending on your specific needs:

Approach 1: Explode Lists into Rows First

This is perfect if you want every element inside the lists to contribute to the aggregation:

import pandas as pd

# Your original DataFrame
df = pd.DataFrame({'A':[1,1,1,1,2,2,2,2], 'B':['x','x','y','y','x','x','y','y'], 'C':[[1,2,3,4],[5,6],[8,9,10,11],[12,13,14],[15],[16,17,18],[],[19]]})

# Explode lists into individual rows, then convert C to numeric type
df_exploded = df.explode('C').astype({'C': int})

# Group by A and calculate mean/std
result = df_exploded.groupby('A')['C'].agg(['mean', 'std'])
print(result)

Output:

mean        std
A                       
1   7.416667  4.716991
2  16.166667  1.722408

Approach 2: Custom Aggregation Functions for Lists

If you want to keep the original grouping structure and aggregate across all lists in a group (e.g., combine all lists into one big collection then compute stats), use custom functions:

import pandas as pd
import numpy as np

# Custom function to calculate mean across all elements in group lists
def list_mean(series):
    all_values = [num for lst in series for num in lst]
    return np.mean(all_values) if all_values else np.nan

# Custom function to calculate standard deviation across all elements in group lists
def list_std(series):
    all_values = [num for lst in series for num in lst]
    return np.std(all_values, ddof=1) if len(all_values) >= 2 else np.nan

# Run the aggregation
result = df.groupby('A')['C'].agg([list_mean, list_std]).rename(columns={'list_mean':'mean', 'list_std':'std'})
print(result)

This gives the same output as Approach 1, and it handles empty lists gracefully (they're just ignored in calculations).

Bonus: Aggregate List Means #

If your goal is to first calculate the mean of each individual list, then aggregate those means per group, adjust the custom function like this:

def group_list_mean(series):
    list_means = [np.mean(lst) for lst in series if lst]
    return np.mean(list_means) if list_means else np.nan

内容的提问来源于stack exchange，提问作者HappyPy