如何在Pandas中合并日期与时间列生成标准时间戳？

Hey there! No worries—combining separate Date and Time columns into a standard timestamp like 2015-07-04 00:00:00 is totally doable with Pandas, and it's simpler than you might think. Let's break this down with your specific dataset.

Step 1: Understand Your Data Format #

First, let's recap your data structure:

• Date uses the YYYY.MM.DD format (e.g., 2000.12.22)
• Time uses the HH:MM format (e.g., 12:00)
• Both columns are pandas.core.series.Series objects

Step 2: Combine and Convert to Timestamp #

We'll use Pandas' pd.to_datetime() function, which excels at parsing date-time strings into proper datetime objects. Here's how to implement this:

First, replicate your sample data (to match your scenario):

import pandas as pd

# Your dataset as provided
data = {
    'Date': ['2000.12.22', '2000.12.22', '2000.12.22', '2000.12.23', '2000.12.25'],
    'Time': ['12:00', '16:00', '20:00', '00:01', '00:00'],
    'Open': [0.91810, 0.92260, 0.92300, 0.92410, 0.92300],
    'High': [0.92620, 0.92520, 0.92580, 0.92450, 0.92460],
    'Low': [0.91650, 0.92220, 0.92260, 0.92270, 0.92300],
    'Close': [0.92320, 0.92310, 0.92420, 0.92320, 0.92420],
    'Vol': [2244, 1688, 955, 168, 260]
}
df = pd.DataFrame(data)

Option 1: Auto-detect the format

Pandas is usually smart enough to recognize the date-time structure on its own. Just concatenate the Date and Time columns with a space, then pass the result to pd.to_datetime():

# Create a new 'Datetime' column with the merged timestamp
df['Datetime'] = pd.to_datetime(df['Date'] + ' ' + df['Time'])

Option 2: Specify the format (for maximum reliability)

If you want to eliminate any auto-detection ambiguity, explicitly define the format using strftime codes:

# %Y = 4-digit year, %m = 2-digit month, %d = 2-digit day
# %H = 24-hour format hour, %M = minute
df['Datetime'] = pd.to_datetime(df['Date'] + ' ' + df['Time'], format='%Y.%m.%d %H:%M')

Step 3: Verify the Result #

Check the new column to confirm it's in your desired format:

print(df['Datetime'])

You'll get output like this:

0   2000-12-22 12:00:00
1   2000-12-22 16:00:00
2   2000-12-22 20:00:00
3   2000-12-23 00:01:00
4   2000-12-25 00:00:00
Name: Datetime, dtype: datetime64[ns]

Optional: Remove Original Columns #

If you no longer need the separate Date and Time columns, drop them to clean up your DataFrame:

df = df.drop(['Date', 'Time'], axis=1)

Why This Works #

The pd.to_datetime() function converts string inputs into Pandas' datetime64[ns] type, which is optimized for time-series tasks like filtering by date, resampling data, or calculating time differences.

Hope this solves your problem! Feel free to ask if you need help with any follow-up steps.

内容的提问来源于stack exchange，提问作者Flying pig