Pandas日期时间列解析求助：自定义编码转标准格式

Got it, let's fix this datetime parsing issue step by step. The core problem with your initial code is that it didn't account for the sequential day/time logic behind those numeric values—you can't parse them directly with a simple %d format string. Here's a working, straightforward approach:

Step 1: Convert columns to numeric types #

First, we need to turn the object-type Date and Time columns into integers so we can calculate proper time offsets:

import pandas as pd

# Convert object columns to integers (handle any parsing errors gracefully if needed)
df['Date'] = df['Date'].astype(int)
df['Time'] = df['Time'].astype(int)

Step 2: Define the base date for your Date sequence #

Since 195 maps to 01/01/2009, we need to find the starting date of the sequence (the date that would correspond to Date=1). We calculate this by subtracting 194 days from 01/01/2009 (because 195 is 194 days after the base date):

base_date = pd.to_datetime('2009-01-01') - pd.Timedelta(days=195 - 1)

Step 3: Map numeric Date values to actual dates #

Now we can convert each numeric Date value to its corresponding real date by adding the day offset to our base date:

# Subtract 1 because base_date is day 1; adding 194 days gives us 01/01/2009 for Date=195
df['date_part'] = base_date + pd.to_timedelta(df['Date'] - 1, unit='D')

Step 4: Map numeric Time values to time offsets #

Your Time column uses 30-minute intervals, where 1 = 00:00:00–00:29:59. Based on your example (wanting 00:30 for Time=1), you're targeting the end time of each interval. We calculate the time offset like this:

# For end time of the interval (matches your example: 1 → 00:30)
df['time_offset'] = pd.to_timedelta(df['Time'] * 30, unit='m')

# Uncomment below if you want the start time instead (1 → 00:00)
# df['time_offset'] = pd.to_timedelta((df['Time'] - 1) * 30, unit='m')

Step 5: Combine and format the final datetime #

Merge the date part and time offset to get a full datetime column, then format it to your preferred string format:

# Create the full datetime column
df['datetime'] = df['date_part'] + df['time_offset']

# Format to your desired string (e.g., "01,01,2009 00:30")
df['datetime_formatted'] = df['datetime'].dt.strftime('%d,%m,%Y %H:%M')

Example Output #

For a row with Date=195 and Time=1, this will produce:

• date_part: 2009-01-01
• time_offset: 0 days 00:30:00
• datetime: 2009-01-01 00:30:00
• datetime_formatted: 01,01,2009 00:30

Which perfectly matches your expected result!

内容的提问来源于stack exchange，提问作者MAHDl EMAD