create python array from dict of dicts

Question

new one for me -- I'm assuming this is fairly easy, but I've never worked with arrays before, so I'm curious as to how it works.

I have a dict of dicts, like so:

{'bob': {'a':1, 'b':2, ...}, 'joe': {'a':2, 'c':3, ...} ...}

I'd like to turn it into an array so I can write it to a CSV and then turn it into a heatmap using R. I was trying to cheat and just write each nested dict to an individual row, but of course that won't work because not every key is present in every nested dict. simple, right?

desired output would look like (in tabular form):

,a,b,c
bob,1,2,0
joe,2,0,3

You have put a sample of your input. Can you give an example representation of the desired output? — Joseph Victor Zammit
– Joseph Victor Zammit, Commented Dec 1, 2012 at 22:18
Your sample output at present does not explicitly address how to handle the case where a key is absent from one of the nested dictionaries. Do you want us to treat its value as zero? — Mark Amery
– Mark Amery, Commented Dec 1, 2012 at 22:22

rhettg · Accepted Answer · 2012-12-01 22:24:28Z

3

If your columns are fixed, you could simply do something like:

cols = ['a', 'b', 'c']
csv.writerow([''] + cols)
for name, values in data.iteritems():
    csv.writerow([name] + [values.get(c, 0) for c in cols])

answered Dec 1, 2012 at 22:24

rhettg

2,45318 silver badges17 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

Sean Vieira Over a year ago

+1 - there is also csv.DictWriter from the stdlib

serilain Over a year ago

okay -- and I could presumably get the names of all of the columns before this step by doing for q,c in dictofdicts.iteritems(): coltitles.append(c.keys()) coltitles = list(set(coltitles))

Samy Arous · Accepted Answer · 2012-12-01 22:26:03Z

1

let's suppose you have 3 predefined keys, you can use the get function of the dict to get the value or a default one if the key is not in the dict:

headers = ('a', 'b', 'c')
for key, values in dict.item():
     print ','.join([values.get(h, '') for h in headers])

answered Dec 1, 2012 at 22:26

Samy Arous

6,81215 silver badges20 bronze badges

Comments

Thomas · Accepted Answer · 2012-12-01 22:40:31Z

1

Others have already answered the printing, but assume fixed headers. To get the column headers from the dict:

columns = sorted(set(column for subdict in dict_of_dicts.itervalues() for column in subdict))

Or, more verbosely:

column_set = set()
for subdict in dict_of_dicts.itervalues():
  for column in subdict:
    column_set.add(column)
columns = sorted(column_set)

To create the array in one long line, just for fun, not recommended:

array = [[''] + columns] + [[key] + [subdict.get(column, 0) for column in columns] for key, subdict in dict_of_dicts.iteritems()]

edited Dec 1, 2012 at 22:40

answered Dec 1, 2012 at 22:34

Thomas

183k57 gold badges383 silver badges510 bronze badges

Collectives™ on Stack Overflow

create python array from dict of dicts

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

2 Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related