CODEWITHSUNDEEP
pythonpython-2.7pandas
staten12
staten12

Reputation: 785

How do I iterate through a pandas group and check to see if a string is in each group?

I have a pandas dataframe with a bunch of records on certain dates. I need to group them by date, and do a check to see if the next day has records that are also in the prior day, specifically I need to output what records were deleted.

Here is an example dataset:

Date        Item
20160101    apple
20160101    pear
20160101    banana
20160102    apple
20160102    pear
20160102    beans

I need to figure out the differences that occur for each date, so for this example from 01/02/2016 there is a an added string 'beans' and a 'banana' was removed from the group.

So far I have as my code:

groups = frame['Item'].groupby(frame['Date'])

for date, item in groups:
    for i in item:
        if i not in item[:-1]:
            print date, item, 'Deleted'

This doesn't seem to be working. I should be expecting:

20160102 , banana, Deleted

Thanks for your help!

Upvotes: 3

Views: 201

Answers (1)

piRSquared
piRSquared

Reputation: 294488

diffs = frame.groupby(frame.columns.tolist()).size().unstack(fill_value=0).diff()
diffs

enter image description here

diffs.mask(diffs.eq(0)).stack().map({-1: 'deleted', 1: 'added'})

Date      Item  
20160102  banana    deleted
          beans       added
dtype: object

Upvotes: 3

Related Questions