使用pd.cut分组DataFrame时出现Grouper与轴长度不一致的ValueError求助

Ah, I’ve run into this exact gotcha before! Let’s figure out what’s going wrong here.

The Root Cause #

When you use pd.cut() with the retbins=True parameter, it doesn’t just return your binned labels—it returns a tuple containing two items:

1. The binned label Series (which matches the length of your original df['time'] column)
2. The actual bin boundaries array (which has length 8 in your case, since you defined 7 bins)

When you pass this entire tuple directly to groupby(), Pandas gets confused because the second element of the tuple (the bins array) has a different length than your DataFrame’s rows. That’s exactly why you’re seeing the ValueError: Grouper and axis must be same length error.

The Fixes #

There are two simple ways to resolve this, depending on whether you need to keep the returned bin boundaries or not:

1. If you need to retain the bin boundaries

Split the tuple returned by pd.cut() into separate variables, so you only pass the binned labels to groupby():

# Split the returned tuple into binned data and bins
frameddata, returned_bins = pd.cut(df['time'], bins, retbins=True, labels=timeframe)
# Now group using only the frameddata Series
groups = df.groupby(frameddata)

2. If you don’t need the bin boundaries

Just remove the retbins=True parameter entirely. pd.cut() will then only return the binned label Series, which matches your DataFrame’s length perfectly:

# No retbins=True, so frameddata is just the binned labels
frameddata = pd.cut(df['time'], bins, labels=timeframe)
# Grouping works as expected
groups = df.groupby(frameddata)

Quick Bonus Tip #

If your df['time'] has values outside the [3, 24] range, pd.cut() will default to setting those values to NaN. If you want to include these NaN values as a separate group in your results, add the dropna=False parameter to groupby():

groups = df.groupby(frameddata, dropna=False)

内容的提问来源于stack exchange，提问作者user3194861