python - Pandas group-by error duplicate axis but no duplicate values

Question

I've done this code:

df[['GL','Libelle']]=df['index'].str.split(' ',1,expand=True)

# Sort by GL, Date
df.sort_values(by=['GL', 'Class','month'], inplace=True)

# add columun with diff by month
df['value'] = pd.to_numeric(df['value'])
df["diff"] = df.groupby(['GL','Class','month'])['value'].diff().fillna(df['value'])

my pandas df is like this: index object

Class object

value float64

glid object

month object

GL object

Libelle object

and this is sample:

Could you explain with I have this error ? "cannot reindex from a duplicate axis" on line df["diff"] = df.groupby(['GL','Class','month'])['value'].diff().fillna(df['value'])

question from:https://stackoverflow.com/questions/65943678/pandas-group-by-error-duplicate-axis-but-no-duplicate-values

Answer 1

Generally, this error arises when you try to join or assign to a column when the index(row or column names) has duplicate values. If I'm understanding correctly, you are trying to join the column, so check if your rows have duplicate values. Also, check your original dataframe. It may possible that the duplicate is present in your original dataframe.

To find the duplicates in the original index, do this: df[df.index.duplicated()]

If you accidentally have created a duplicate column then remove it. This will solve your error.