如何在DataFrame中添加ID列？按数值区间分配对应ID值

Hey there! Let's tackle this problem step by step. You want to add an ID column to your DataFrame where IDs are set to 1 if the numeric value falls between 0 and 100, and 2 otherwise. I'll walk you through how to do this using pandas (the standard library for DataFrames in Python), including handling your provided sample data.

First, let's turn your sample data into a proper pandas DataFrame. Here's the code to replicate your dataset:

import pandas as pd
import numpy as np

# Your sample data converted into a dictionary
sample_data = {
    'timestamp': [
        '2018-03-19 14:31:17.200',
        '2018-03-19 14:31:17.300',
        '2018-03-19 14:31:17.400',
        '2018-03-19 14:31:17.500',
        '2018-03-19 14:31:17.600',
        '2018-03-19 14:31:17.700',
        '2018-03-19 14:31:17.800'
    ],
    'value': [1095, 2296, 2147, 309, 244, 263, 548]
}

# Create the DataFrame
df = pd.DataFrame(sample_data)
# Optional (but recommended): Convert timestamp to datetime type
df['timestamp'] = pd.to_datetime(df['timestamp'])

Now, let's add the ID column using two common methods—pick the one that fits your workflow best:

Method 1: Using numpy.where() (Fastest for Large Datasets) #

This is the most efficient approach, especially if you're working with big data. It vectorizes the condition check instead of processing rows one by one:

# Add ID column: 1 if value is 0-100, else 2
df['ID'] = np.where((df['value'] >= 0) & (df['value'] <= 100), 1, 2)

Method 2: Using pandas.apply() (More Explicit for Simple Logic) #

If you prefer a readable, row-by-row approach (great for small datasets or when you need to expand logic later), use apply:

def get_id(row):
    if 0 <= row['value'] <= 100:
        return 1
    else:
        return 2

df['ID'] = df.apply(get_id, axis=1)

Final Result #

After running either method, your DataFrame will look like this (note all sample values are above 100, so all IDs are 2):

Quick tip: If you need to adjust the range (like excluding 0 or 100), just tweak the condition in either method—super easy to modify!

内容的提问来源于stack exchange，提问作者Maphanew Kim