Renaming the index in a pandas Series

user_17051433 0
Viewed 67

For the past few hours, I've been trying to change/rename the name of my multiindex in my pandas Series but I do not know how to achieve it. I've been looking all kind of different things and I cannot seem to get it right. This is important because I have realized that if I do not do so, my two indexes have the same name (which is "date"). Here is the code:

df_bar = df.groupby([pd.Grouper(freq = "Y"), pd.Grouper(freq = "M")])["value"].mean()

if you print df_bar, you would have the following: print of df_bar

Would it be possible to name each index directly in the groupby()?

Thank you in advance for the help :)

pandas series dataframe visualization python
1 Answers

I think not, need rename in next step by Series.rename_axis:

df = pd.DataFrame(
   {
       "date": [
            pd.Timestamp("2000-01-02"),
            pd.Timestamp("2000-01-02"),
            pd.Timestamp("2001-01-09"),
            pd.Timestamp("2001-01-16")
        ],
        "value": [10, 20, 30, 40]
    }

).set_index('date')

df_bar = (df.groupby([pd.Grouper(freq = "Y"), pd.Grouper(freq = "M")])["value"]
            .mean()
            .rename_axis(['Y','M']))
print (df_bar)
Y           M         
2000-12-31  2000-01-31    15
2001-12-31  2001-01-31    35
Name: value, dtype: int64
user_29010020
Related