多气象站小时温度DataFrame的多时间尺度均值循环实现方法咨询

Great question—repeating the same resampling code for every station is tedious, so let's fix that with a clean, reusable workflow. Here's how to do it efficiently:

First, Prep Your Data #

Make sure every station's DataFrame has a datetime index (or a column you can convert to one). This is non-negotiable for pandas resampling. For example:

# For a single station's DataFrame
import pandas as pd
station_df['timestamp'] = pd.to_datetime(station_df['timestamp'])
station_df.set_index('timestamp', inplace=True)

Do this for all your station datasets first—consistency here will make the rest smooth.

Step 1: Write a Reusable Aggregation Function #

Instead of copying resampling code for each station, define one function that handles all the temporal averages. This is where you centralize your logic:

def calculate_temporal_means(df, temp_column='temperature'):
    # Compute all resampled averages in one go
    daily = df[temp_column].resample('D').mean().rename('daily_avg')
    weekly = df[temp_column].resample('W').mean().rename('weekly_avg')
    monthly = df[temp_column].resample('M').mean().rename('monthly_avg')
    annual = df[temp_column].resample('Y').mean().rename('annual_avg')
    
    # Combine into a single DataFrame for easy access
    return pd.concat([daily, weekly, monthly, annual], axis=1)

You can tweak the resampling codes (like 'W-MON' for weeks starting on Monday) or add more intervals right here if needed.

Step 2: Loop Through All Stations #

Store your station DataFrames in a dictionary (keys = station names, values = DataFrames) for easy iteration. Then loop through and apply the function:

# Example: Dictionary holding all station data
station_datasets = {
    'Downtown': downtown_df,
    'Uptown': uptown_df,
    'Suburb': suburb_df
}

# Dictionary to store results for each station
station_averages = {}

# Loop and compute averages
for station_name, df in station_datasets.items():
    print(f"Processing {station_name}...")
    station_averages[station_name] = calculate_temporal_means(df)

# Access results for a specific station
print(station_averages['Downtown']['daily_avg'].head())

Optional: Combine Results for Cross-Station Comparison #

If you want to compare averages across stations, merge all results into a multi-index DataFrame:

combined_results = pd.concat(station_averages.values(), keys=station_averages.keys(), axis=1)

# Example: View daily averages for all stations
print(combined_results.xs('daily_avg', level=1, axis=1).head())

Quick Tips #

• Handle Missing Data: Add dropna=False to resample() if you want to keep periods with no data, or use df.fillna(method='ffill') before resampling to fill gaps (use cautiously!).
• Efficiency: Pandas resampling is optimized, so even with large datasets, this loop will run quickly—you're just applying pre-built logic to each dataset without redundant code.
• Scalability: Add new stations to the station_datasets dictionary, and the loop will handle them automatically without any extra code.

内容的提问来源于stack exchange，提问作者Antonio Bonilla