Argthtjtr

Currently, I have a dictionary, with its key representing a zip code, and the values are also a dictionary.

d = { 94111: {'a': 5,  'b': 7,  'd': 7}, 

      95413: {'a': 6,  'd': 4}, 

      84131: {'a': 5,  'b': 15, 'c': 10, 'd': 11}, 

      73173: {'a': 15, 'c': 10, 'd': 15}, 

      80132: {'b': 7,  'c': 7,  'd': 7} }

And then a second dictionary, which associates which state the zip code belongs to.

states = {94111: "TX", 84131: "TX", 95413: "AL", 73173: "AL", 80132: "AL"}

If the zip code in the dictionary states matches one of the keys in db then it would sum up those values and put it into a new dictionary like the expected output.

Expected Output:

{'TX': {'a': 10, 'b': 22, 'd': 18, 'c': 10}, 'AL': {'a': 21, 'd': 26, 'c': 17, 'b': 7}}

I've been stuck on how to proceed for almost an hour now and have absolutely no clue. So far this is the direction I am looking to go into but i'm not sure when both the keys match, how to create a dictionary that will look like the expected output.

def zips(d, states):

    result = dict()

    for key, value in db.items():

        for keys, values in states.items():

            if key == keys:





zips(d, states)

Using collections module

Ex:

from collections import defaultdict, Counter



d = { 94111: {'a': 5,  'b': 7,  'd': 7}, 

      95413: {'a': 6,  'd': 4}, 

      84131: {'a': 5,  'b': 15, 'c': 10, 'd': 11}, 

      73173: {'a': 15, 'c': 10, 'd': 15}, 

      80132: {'b': 7,  'c': 7,  'd': 7} }



states = {94111: "TX", 84131: "TX", 95413: "AL", 73173: "AL", 80132: "AL"}



result = defaultdict(Counter)

for k,v in d.items():

    if k in states:

        result[states[k]] += Counter(v)

print(result)

Output:

defaultdict(<class 'collections.Counter'>, {'AL': Counter({'d': 26, 'a': 21, 'c': 17, 'b': 7}), 

'TX': Counter({'b': 22, 'd': 18, 'a': 10, 'c': 10})})

You can just use defaultdict and count in a loop:

expected_output = defaultdict(lambda: defaultdict(int))

for postcode, state in states.items():

     for key, value in d.get(postcode, {}).items():

         expected_output[state][key] += value

Just as a complement of the answer of Rakesh, Here is an answer closer to your code:

res = {v:{} for v in states.values()}



for k,v in states.items():

    if k in d:

        sub_dict = d[k]

        output_dict = res[v]

        for sub_k,sub_v in sub_dict.items():

            output_dict[sub_k] = output_dict.get(sub_k, 0) + sub_v

You can use something like this:

d = { 94111: {'a': 5,  'b': 7,  'd': 7},                                                                                                                                                

      95413: {'a': 6,  'd': 4},                                                                 

      84131: {'a': 5,  'b': 15, 'c': 10, 'd': 11},                                              

      73173: {'a': 15, 'c': 10, 'd': 15},                                                       

      80132: {'b': 7,  'c': 7,  'd': 7} }                                                       

states = {94111: "TX", 84131: "TX", 95413: "AL", 73173: "AL", 80132: "AL"}                         



out = {i: 0 for i in states.values()}                                                              

for key, value in d.items():                                                                       

    if key in states:                                                                              

        if not out[states[key]]:                                                                   

            out[states[key]] = value                                                               

        else:                                                                                      

            for k, v in value.items():                                                             

                if k in out[states[key]]:                                                          

                    out[states[key]][k] += v                                                       

                else:                                                                              

                    out[states[key]][k] = v                                                        

# out -> {'TX': {'a': 10, 'b': 22, 'd': 18, 'c': 10}, 'AL': {'a': 21, 'd': 26, 'c': 17, 'b': 7}}

You can use the class Counter for counting objects:

from collections import Counter



d = { 94111: {'a': 5,  'b': 7,  'd': 7}, 

      95413: {'a': 6,  'd': 4}, 

      84131: {'a': 5,  'b': 15, 'c': 10, 'd': 11}, 

      73173: {'a': 15, 'c': 10, 'd': 15}, 

      80132: {'b': 7,  'c': 7,  'd': 7} }



states = {94111: "TX", 84131: "TX", 95413: "AL", 73173: "AL", 80132: "AL"}



new_d = {}

for k, v in d.items():

    if k in states:

        new_d.setdefault(states[k], Counter()).update(v)



print(new_d)

# {'TX': Counter({'b': 22, 'd': 18, 'a': 10, 'c': 10}), 'AL': Counter({'d': 26, 'a': 21, 'c': 17, 'b': 7})}

You can convert new_d to the dictionary of dictionaries:

for k, v in new_d.items():

    new_d[k] = dict(v)



print(new_d)

# {'TX': {'a': 10, 'b': 22, 'd': 18, 'c': 10}, 'AL': {'a': 21, 'd': 26, 'c': 17, 'b': 7}}

You can leverage dict's .items() method, which returns a list of tuples, and get the expected output in a simple one-liner:

new_dict = {value:d[key] for key, value in states.items()}

Output:

{'AL': {'b': 7, 'c': 7, 'd': 7}, 'TX': {'a': 5, 'b': 15, 'c': 10, 'd': 11}}

You might want to reconsider your choice of dict for how to store your data. If you store your data using pandas, aggregation is a lot easier.

df = pd.DataFrame(d).transpose()

df['states']=pd.Series(states)

df.groupby('states').sum()



>>            a     b     c     d

>>states                        

>>AL      21.0   7.0  17.0  26.0

>>TX      10.0  22.0  10.0  18.0