0

I was looking for a way to find the mean of numerical values based on a certain column. I looked to this link for advice but it requires all values of one column to be the same and I'm sure there's a Pythonic way that would do that for all values that are duplicates.

Here's an example.

data = {
  "Name": ["John", "John", "Robert", "Robert", "Cindy", "Cindy", "Sarah", "Sarah"],
  "Score": [84, 45, 67, 87, 88, 100, 76, 91]
}

#load data into a DataFrame object:
df = pd.DataFrame(data)

df

enter image description here

I'd like it so there's one row of John with whatever the mean of John is. Same with Robert, Cindy and Sarah.

Thanks!

E_Sarousi
  • 151
  • 9

2 Answers2

1
# groupby and mean
df.groupby('Name', as_index=False)['Score'].mean()

Name    Score
0   Cindy   94.0
1   John    64.5
2   Robert  77.0
3   Sarah   83.5
Naveed
  • 11,495
  • 2
  • 14
  • 21
1
new_df = df.groupby(['Name'], as_index=False).agg({'Score': pd.Series.mean})

     Name  Score
0   Cindy   94.0
1    John   64.5
2  Robert   77.0
3   Sarah   83.5
andrew
  • 81
  • 1
  • 8