Bin and aggregate with `seaborn.objects`

Question

Is there a way to both bin and aggregate (with some other function than count) in seaborn.objects? I'd like to compute the mean per bin and right now I'm using the following:

import seaborn.objects as so
import pandas as pd
import seaborn as sns

df = sns.load_dataset("penguins")
df2 = (
    df.groupby(pd.cut(df["bill_length_mm"], bins=30))[["bill_depth_mm"]]
    .mean()
)
df2["bill_length_mm"] = [x.mid for x in df2.index]
p = so.Plot(df2, x="bill_length_mm", y="bill_depth_mm").add(so.Bars())

p

tdy · Accepted Answer · 2023-02-07 17:23:33Z

3

There's not yet a binning operation separate from Hist (this could make sense as a Stat or a Scale, I'm not sure).

But note that you can do the aggregation more simply than you are in your example because you can pass a Series directly and don't need to construct a new dataframe:

(
    so.Plot(
        df,
        x="bill_depth_mm",
        y=pd.cut(df["bill_length_mm"], bins=30).map(lambda x: x.mid),
    )
    .add(so.Bars(), so.Agg("mean"))
)

Note that the Series will be aligned with the DataFrame (or other Series passed directly) based on the index information rather than position.

edited Feb 7, 2023 at 17:23

tdy

42.1k42 gold badges124 silver badges125 bronze badges

answered Feb 7, 2023 at 12:08

mwaskom

49.3k16 gold badges135 silver badges137 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

nicolaskruchten Over a year ago

Thanks! "Note that the Series will be aligned with the DataFrame (or other Series passed directly) based on the index information rather than position." ... I'm not sure what this means :)

tdy Over a year ago

e.g., s1 = pd.Series([1, 2, 3], index=[*'ABC']) and s2 = pd.Series([1, 2, 3], index=[*'BCA']) would align as 1:3, 2:1, 3:2 (based on A, B, C) rather than 1:1, 2:2, 3:3 (based on position in array)

Collectives™ on Stack Overflow

Bin and aggregate with `seaborn.objects`

1 Answer 1

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related