Appending groupbys to a nested dictionary

Question

Quick summary of my problem: ghGby is a dictionary of keys each corresponding to a dataframe groupby. I would like to divide each groupby into their own groupby's corresponding to each set of days in the column of the groupby.

groupDays = {}
groupDayIndexer = []
for x in ghGby.keys():
    for y in ghGby[x].DAY.unique().keys():
        if x in groupDays:
            groupDays[x].append(ghGby[x].get_group(y))
        else:
            groupDays[x] = ghGby[x].get_group(y)

Data looks like this:

ghGby[keyA]:

day|transaction|etc.
51 | ......... | ...
51 | ......... | ...
63 | ......... | ...
63 | ......... | ...
63 | ......... | ...
94 | ......... | ...

.get_group(y) returns each set of days as an individual object just fine, but when I append them to groupDays I only get one day of the groupby rather than each one like this:

print(groupDays['keyA'])
{keyA: [day51GroupBy, day63GroupBy, day94GroupBy]}

more background information:

original dataset looks like this, just many thousands of household_keys. My objective is to be able to access a subset of this large dataset by specifying my desired day and desired household key. As these are transactions, the same key can have multiple entries on the same day.

household_key   DAY  PRODUCT_ID
1929            4    1004906
1929            4    1004906
1929            95   1004906
1929            202  1004906
1929            207  1004906

my desired output:

print(groupDays['household_key1929'])
{[ghGby[groupDays['household_key1929'].get_group(day4), ghGby[groupDays['household_key1929'].get_group(day52), ghGby[groupDays['household_key1929'].get_group(day95), ghGby[groupDays['household_key1929'].get_group(day202)]}

I would like to do this so that I can access my data easier, like this:

display(groupDays['household_key1929'][0])

household_key   DAY  PRODUCT_ID
1929            4    1004906
1929            4    1004906

I am accessing the first element of the list of days associated to household key 1929, in this case it would be day 4.

please, could you please show us the original dataset? and the desired output as well. Thanks! — Samir Hinojosa, Jun 09 '21 at 08:40
@SamirHinojosa hello Samir- just updated with some more information, hope it makes things clearer — Bigboss01, Jun 09 '21 at 09:03
Could you please to show us the desired output as a dataset? Thanks! — Samir Hinojosa, Jun 09 '21 at 09:08
I don't know what is the data in dict's values... but I guess you need to work with [`explode()`](https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.explode.html), you can see a [`related answer`](https://stackoverflow.com/questions/53218931/how-to-unnest-explode-a-column-in-a-pandas-dataframe) and after that make a `groupby()` based on yours needs — Samir Hinojosa, Jun 09 '21 at 09:28

score 0 · Accepted Answer · answered Jul 01 '21 at 15:04

simple nested dictionary:

dic2make = {}
for indexer in dataList:
    dic2make[indexer] = {}
    for element in groupbyDict[indexer].DAY.unique().keys():
        dic2make[indexer][element] = groupbyDict[indexer].get_group(element)

Very simple solution to a simple question. Am beginner therefore my description of the problem was very confusing. Anyways- this is what I was looking for. If none of this makes sense to you but you think you need to understand what's happening, feel free to leave a comment.

Appending groupbys to a nested dictionary

1 Answers1