filter list of object with multiple condition in python

Question

i have list structure look like this :

example =
[
   {
      "value":"promo",
      "score":0.3333333333333333,
      "slugger":"promoKeyword",
      "type":"normal",
   },
   {
      "value":"unknown",
      "score":1.0,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.3333333333333333,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.5,
      "slugger":"promoCart",
      "type":"normal",
   }
]

i want to filter the list by maximum score in [score] key if only the [slugger] key has same value(this mean [slugger] can have multiple same value and we only take the highest score of it)

so the example will look like this

[
   {
      "value":"promo",
      "score":0.3333333333333333,
      "slugger":"promoKeyword",
      "type":"normal",
   },
   {
      "value":"unknown",
      "score":1.0,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.5,
      "slugger":"promoCart",
      "type":"normal",
   }
]

my effort right now look like this,but it fails to satisfied the condition

score_data = []
for data in example:
    score_data.append(data['score'])
max_score = max(score_data)
example = [x for x in example if x['score'] == max_score and x['score'] > 0]
example = list({ each['slug'] : each for each in example }.values())

can you guys help ? thank you in advance..pardon my english

I don't have much time, so only general advice. Read about groupby - sort by slugger value, then group by it (groupby only groups adjacent elements, hence the sorting first), and then you can take the max. — h4z3
– h4z3, Commented Nov 27, 2019 at 16:21

Andrej Kesely · Accepted Answer · 2019-11-27 16:25:04Z

1

One solution using itertools:

data = [
   {
      "value":"promo",
      "score":0.3333333333333333,
      "slugger":"promoKeyword",
      "type":"normal",
   },
   {
      "value":"unknown",
      "score":1.0,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.3333333333333333,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.5,
      "slugger":"promoCart",
      "type":"normal",
   }
]

from itertools import groupby, islice

rv = []
for _, g in groupby(sorted(data, key=lambda k: (k['slugger'], -k['score'])), lambda k: k['slugger']):
    rv.extend(islice(g, 0, 1))

from pprint import pprint
pprint(rv, width=30)

Prints:

[{'score': 0.5,
  'slugger': 'promoCart',
  'type': 'normal',
  'value': 'theory'},
 {'score': 1.0,
  'slugger': 'promoCategory',
  'type': 'normal',
  'value': 'unknown'},
 {'score': 0.3333333333333333,
  'slugger': 'promoKeyword',
  'type': 'normal',
  'value': 'promo'}]

answered Nov 27, 2019 at 16:25

Andrej Kesely

196k15 gold badges60 silver badges105 bronze badges

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

1 Comment

Androidnoob Over a year ago

thank you!...solved ,its really helpfull..i need to explore itertools then..thanks again

Vash · Accepted Answer · 2019-11-27 16:22:08Z

Perhaps convert the list of dictionaries to a dataframe and then extract the stuff you want?

list_values = [
   {
      "value":"promo",
      "score":0.3333333333333333,
      "slugger":"promoKeyword",
      "type":"normal",
   },
   {
      "value":"unknown",
      "score":1.0,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.3333333333333333,
      "slugger":"promoCategory",
      "type":"normal",
   },
   {
      "value":"theory",
      "score":0.5,
      "slugger":"promoCart",
      "type":"normal",
   }
]

df = pd.DataFrame(list_values)

# Get average scores for each slugger:
df.groupby('slugger')['score'].mean()

# Get max score for each slugger:
df.groupby('slugger')['score'].max()

You haven't specified what the example variable is, so I can't really help you with that.

maede rayati · Accepted Answer · 2019-11-27 16:48:24Z

0

You can fist create the filter feature dictionary and then create a new list based on this filter dictionary. For example in your example the code will look like this.

d = dict()

## this will create a dictionary of categories as keys and highest score as value

for e in example:
   if e['slugger'] in d:
     if e['score']> d['slugger']:
       d['slugger'] = e['score']
   else:
     d[e['slugger']] = e['score']

## this will filter the original list by dictionary
result = [e for e in example if d[e['slugger']] == e['score']]

answered Nov 27, 2019 at 16:48

maede rayati

7866 silver badges11 bronze badges

Comments

Felipe Endlich · Accepted Answer · 2019-11-27 16:52:54Z

Use list comprehensions

data = [
    {
        "value":"promo",
        "score":0.3333333333333333,
        "slugger":"promoKeyword",
        "type":"normal",
    },
    {
        "value":"unknown",
        "score":1.0,
        "slugger":"promoCategory",
        "type":"normal",
    },
    {
        "value":"theory",
        "score":0.3333333333333333,
        "slugger":"promoCategory",
        "type":"normal",
    },
    {
        "value":"theory",
        "score":0.5,
        "slugger":"promoCart",
        "type":"normal",
    }]
print([
    max([y['score'] for y in data if y['slugger'] == x]) 
        for x in set([z['slugger'] for z in data])
])

set([z['slugger'] for z in data])

That part creates an iterable element with unique values, in your case, unique 'slugger' values.

[[y['score'] for y in data if y['slugger'] == x] for x in set([z['slugger'] for z in data])]

That part return the scores grouped in a list by the sluggers.

And finally we use max to get only the max values of each group.

Collectives™ on Stack Overflow

filter list of object with multiple condition in python

4 Answers 4

1 Comment

Comments

Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

1 Comment

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related