Pandas reset_index(drop=True) 无法与 groupby 一起正常工作

Question

I have a pandas Datarame as below:

data = {
    'user_id': [1, 1, 1, 1, 1, 2],
    'tag_id': [700, 700, 700, 701, 701, 700],
    'score': [1, 0.9, 0.8, 0.7, 1, 0.6]
}
df = pd.DataFrame(data)

>>> df
   user_id  tag_id  score
0        1     700    1.0
1        1     700    0.9
2        1     700    0.8
3        1     701    0.7
4        1     701    1.0
5        2     700    0.6

Then I applied a groupby operation on my DataFrame as below, which basically sums scores in a group, but it doesn't remove indexes :

df = df.groupby(['user_id', 'tag_id'], as_index=False).sum().reset_index(drop=True)

>>> df
   user_id  tag_id  score
0        1     700    2.7
1        1     701    1.7
2        2     700    0.6

I also tried removing as_index parameter, but it removes other columns, and still the indexes are not removed:

df = df.groupby(['user_id', 'tag_id']).sum().reset_index(drop=True)

>>> df
   score
0    2.7
1    1.7
2    0.6

Any ideas on how I can remove these indexes ?

Answer 1

In the comments OP expressed his goal

I want to store this DataFrame without indexes with df.to_sql command into a database. My main problem is on how can I ignore these indexes

In order to do that, passing index=false in pandas.DataFrame.to_sql solves the problem

df.to_sql(index=False)