Function of a formula in python for big dataset-CodePudding

Say, I need to convert the following formula into python and use it on a data frame. The formula is as follows.

The data frame has got values for Xj and Bj. The data frame looks like the following.

 df.head()
type name   OR
1   SAP1    11.21
1   SAP1    1301
2   SAP1    0.7578
2   LUF1    1447
2   LUF1    0.7578
1   ANK3    1150
1   ANK3    0.9909
1   ANK3    1535
1   ACR     0.9909
1   ACR     1535

The above data frame has got values for Xj=type and bj=OR from the data frame. I need for each sam The ^S score from the formula In the end, for each name, I need a single S score.

I have implemented something like this,

   def score(df):
        df_sum   =df[['type','name','OR']].groupby('name').sum().reset_index()
        sum   =df_sum['type']
        OR=df_sum['OR']
        score=sum([sum*OR])
        return score

The question is I need to get them for each name value single score not many scores for all repeating name values. In the end, I need only 4 rows with 4 score values.

It is possible on small datasets. However, when I give a big data frame with multiple name values for OR and type columns. Then I have multiple scores for the same name

Note: The values in column type are binaries. Hence, I am not sure I am use in grouby to count or sum it

CodePudding user response：

I think if you change the order of opérations you can have the expected result:

1 calculate xi*bi

df['xibi'] =df ['type'] * df['OR']

2 groupby and sum

res=df.groupby('name')['xibi'].sum()