Pandas group all rows except selected

user_2140821 0
Viewed 81

I'm trying to do some select grouping and sum but keep getting error messages, clearly this is all wrong. I'd like to group all rows except a few and sum that main group. Basically, I'd like to make a pie chart showing a few slices and the main group as "other". Thanks in advance.

d = {'symbol': ['a', 'b', 'c', 'd', 'e', 'f', 'g'], 'value': [5, 5, 5, 5, 1, 1, 1]}
df = pd.DataFrame(data=d)
display(df)

Output:

    symbol  value
0   a       5
1   b       5
2   c       5
3   d       5
4   e       1
5   f       1
6   g       1

Desired outcome:

    symbol  value
0   rest    20
4   e       1
5   f       1
6   g       1

Is this a somewhat right approach?

def not_in_group(sym):
    if (sym != 'e' or sym != 'f' or sym != 'g' in df['symbol']):
        return sym
    
grouped = df.groupby(not_in_group, axis=1)('value').sum()
display(grouped)
pandas
1 Answers

You can use isin and np.where to map the symbols:

mapped_symbols = np.where(df.symbol.isin(main_groups), df.symbol, 'rest')
df['value'].groupby(mapped_symbols).sum()

Output:

e        1
f        1
g        1
rest    20
Name: value, dtype: int64
user_42384080
Related