熊猫groudby数据框，并获取每组平均值和最常见的值

Question

I have a dataframe with 2 columns.

df=pd.DataFrame({'values':arrays,'ii':lin_index})

I want to group the values by the lin_index and get the mean per group and the most common value per group I try this

bii=df.groupby('ii').median()
bii2=df.groupby('ii').agg(lambda x:x.value_counts().index[0])
bii3=df.groupby('ii')['values'].agg(pd.Series.mode)

I wonder if bii2 and bii3 return the same values Then I want to return the mean and most common value to the original array

bs=np.zeros((np.unique(array).shape[0],1))
bs[bii.index.values]=bii.values

Does this look good?

df looks like

          values        ii
0            1.0  10446786
1            1.0  11316289
2            1.0  16416704
3            1.0  12151686
4            1.0  30312736
     ...       ...
93071038     3.0  28539525
93071039     3.0  19667948
93071040     3.0  22240849
93071041     3.0  22212513
93071042     3.0  41641943

[93071043 rows x 2 columns]

Answer 1

something like this maybe:

# get the mean
df.groupby(['ii']).mean()
# get the most frequent
df.groupby(['ii']).agg(pd.Series.mode)

your question seems similar to GroupBy pandas DataFrame and select most common value

this link might also be useful https://pandas.pydata.org/pandas-docs/stable/reference/frame.html#computations-descriptive-stats