CODEWITHSUNDEEP
pythonoptimization
freethrow
freethrow

Reputation: 1108

Elegant optimization in Python

I have some solutions to this problem, but I am awarre that somewhere out there lies an elegant solution, maybe a two liner.

I have a huge selection (M) of items, basically dictionaries with numerical features like: ItemOne = {width:5, height:10, cost:200,...}

I would like to split this set of dictionaries/items in groups of N (2, 3,...) so that the differences between i.e. width, height or other features should be kept to a minimum according to a criterion (I was thinking a sum of squared differences). The part with the criterion isn't a problem, I just have trouble figuring out the nicest way to split the dataset and get all the combinations without repeating the subsets.

Upvotes: 1

Views: 190

Answers (1)

cmd
cmd

Reputation: 5830

not super clear what you are asking, but I'll give it a shot

#example items
items = [{'width':5, 'height':10, 'cost':200}, {'width':6, 'height':9, 'cost':2}]

#whatever you want your criteria to be
def calculate_criteria(item):
    return item['width']+item['height']+item['cost']

# create subsets based on criterion
subsets = {}
for item in items:
    criteria = calculate_criteria(item)
    subset = subsets.get(criteria, list())
    subset.append(item)
    subsets[criteria] = subset

print subsets
{17: [{'width': 6, 'cost': 2, 'height': 9}], 215: [{'width': 5, 'cost': 200, 'height': 10}]}

or even better using collections.defaultdict

# create subsets based on criterion
subsets = collections.defaultdict(list)
for item in items:
    subsets[calculate_criteria(item)].append(item)

Upvotes: 1

Related Questions