CODEWITHSUNDEEP
pythonpandas
SCool
SCool

Reputation: 3375

Pandas groupby SyntaxError: keyword can't be an expression

I was doing a simple groupby, but got a deprecation / future warning message.

Here's my original groupby, which works 100% OK:


print(df)

   customer_ID  amount
0          joe     321
1          joe       5
2          joe       1
3          joe    3135
4         mary      35
5         mary      65
6         mary      68
7         mary     654
8         mary       1
9         john    5313
10        john      51
11        john     351
12        john      84
13        john      68

df.groupby('customer_ID')['amount'].agg({'count_transactions':'count','median_spend':'median'})

             count_transactions  median_spend
customer_ID                                  
joe                           4           163
john                          5            84
mary                          5            65

FutureWarning: using a dict on a Series for aggregation is deprecated and will be removed in a future version. Use named aggregation instead.

>>> grouper.agg(name_1=func_1, name_2=func_2)

So I decided to try their suggestion, which is grouper.agg(name_1=func_1, name_2=func_2)

But when I try the suggested method I get a SyntaxError:

df.groupby('customer_ID')['amount'].agg('count_transactions'=count,'median_spend'=median)

SyntaxError: keyword can't be an expression

I've tried a few variations of this, such as adding / removing quotations, but can't figure out how to solve it.

I know my first groupby works fine.. but I'd still like to know where I'm going wrong with the suggestion from the future warning message.

Upvotes: 0

Views: 828

Answers (2)

sahinakkaya
sahinakkaya

Reputation: 6056

Since it's a SyntaxError, it can't be related with pandas. You can get the same error with this:

>>> print("Hello", 'end'="")
  File "<stdin>", line 1
SyntaxError: keyword can't be an expression

So the problem is with the quotes around keyword argument. You need to remove them.

You could also get rid of deprecation warning by unpacking your dict with **.

This

df.groupby('customer_ID')['amount'].agg(**{'count_transactions':'count','median_spend':'median'})

will work as same as this one:

df.groupby('customer_ID')['amount'].agg(count_transactions='count',median_spend='median')

Upvotes: 1

SCool
SCool

Reputation: 3375

Basically the functions have to be strings.

I had to add quotes around count and median functions, and remove quotes from new column names count_transactions and median_spend.

Solution:

df.groupby('customer_ID')['amount'].agg(count_transactions='count',median_spend='median')

Upvotes: 0

Related Questions