如何为DataFrame按五年周期及多维度分组计算指定字段均值？

Hey Guillermo, I’ve got you covered on this one! Let’s walk through how to create that new DataFrame with the five-year period averages you need.

Step 1: Add the Five-Year Period Identifier Column #

First, we need to make sure your original table DataFrame has a column that labels each row with its corresponding five-year period.

If you already have your five-year period vector ready, just assign it directly to a new column:

# Replace 'your_five_year_vector' with your actual vector/series
table['FiveYearPeriod'] = your_five_year_vector

If you don’t have the vector mapped yet (e.g., you only have a Year column in your data), you can generate the period labels with a simple function or pd.cut:

# Option 1: Custom function to create human-readable period strings (e.g., "1990-1994")
def assign_five_year_period(year):
    period_start = (year // 5) * 5
    period_end = period_start + 4
    return f"{period_start}-{period_end}"

table['FiveYearPeriod'] = table['Year'].apply(assign_five_year_period)

# Option 2: Use pd.cut for interval-based labels (e.g., "[1990, 1995)")
# Adjust the start/end values to match your data's time range
table['FiveYearPeriod'] = pd.cut(
    table['Year'],
    bins=range(1990, 2021, 5),  # Example: 1990-1994, 1995-1999, ..., 2015-2019
    right=False
).astype(str)  # Convert interval to string for readability

Step 2: Group and Calculate the Averages #

Now we’ll group your data by the required columns plus the new FiveYearPeriod column, then compute the mean for tas_month and pr_month:

new_table = table.groupby(
    ['Month', 'Longitude', 'Latitude', 'Model', 'Scenario', 'FiveYearPeriod'],
    as_index=False  # Keeps group columns as regular columns instead of index
)[['tas_month', 'pr_month']].mean()

Quick Notes #

• The as_index=False parameter is key here—it ensures your result is a flat DataFrame instead of a MultiIndex table, which is probably what you want for further analysis.
• If your data has missing values, you can explicitly skip them in the mean calculation with .mean(skipna=True) (this is the default behavior, but it’s good to be explicit if you want to document it).
• Double-check that all your grouping columns are in the correct format (e.g., Month is an integer or categorical, not a string like "Jan" unless you want to group by month name).

内容的提问来源于stack exchange，提问作者Guillermo.D