如何在Pandas DataFrame中使用elif逻辑创建分类列

Hey there! Let's sort out that category labeling issue you're working through with your pandas DataFrame. The problem with your current if/elif/else approach is that when you run it directly on a pandas Series (like df['Sales']), you'll hit a ValueError—this is because the condition returns a boolean array (one True/False per row) instead of a single boolean value that standard if statements expect.

Pandas thrives on vectorized operations for these kinds of tasks, so here are three reliable solutions to get your desired Category column:

1. Use numpy.select() (Best for Multiple Conditions) #

This method lets you define clear condition-value pairs and a default for all other cases, making it easy to read and scale if you add more categories later.

import numpy as np
import pandas as pd

# First, let's recreate your sample DataFrame (adjust if your actual data differs)
df = pd.DataFrame({
    'Company': ['MC'] * 10,
    'Sales': [360.0, 340.0, 338.5, 335.5, 235.0, 235.0, 234.0, 127.0, 121.0, 120.5]
})

# Define your conditions and matching categories
conditions = [
    df['Sales'] >= 300,
    (df['Sales'] >= 200) & (df['Sales'] < 300)
]
category_labels = ['Fast Mover', 'Medium Fast Mover']

# Assign the categories (default to 'Slow Mover' for all other cases)
df['Category'] = np.select(conditions, category_labels, default='Slow Mover')

2. Use pandas.cut() (Perfect for Binning Numeric Data) #

If your categories are based on continuous numeric ranges, pd.cut() is a clean, concise option. It automatically maps values to predefined bins:

df['Category'] = pd.cut(
    df['Sales'],
    bins=[-float('inf'), 200, 300, float('inf')],  # Define your range boundaries
    labels=['Slow Mover', 'Medium Fast Mover', 'Fast Mover']  # Match bins to labels
)

Note: pd.cut() uses left-closed, right-open intervals by default, so this setup correctly groups values <200 as Slow Mover, 200-299.999 as Medium Fast Mover, and 300+ as Fast Mover.

3. Use apply() (Simple for Small Datasets) #

While not as efficient for large datasets (it processes rows one by one), a custom function with apply() works if you prefer the familiar if/elif structure:

def get_category(sales_value):
    if sales_value >= 300:
        return 'Fast Mover'
    elif 200 <= sales_value < 300:
        return 'Medium Fast Mover'
    else:
        return 'Slow Mover'

df['Category'] = df['Sales'].apply(get_category)

Verify the Result #

All three methods will produce your desired output:

For most cases, I'd recommend numpy.select() or pandas.cut()—they're vectorized operations, so they'll run much faster on large datasets compared to apply().

内容的提问来源于stack exchange，提问作者Ahamed Moosa