Question

我有大约1000个这样的字典。

x = {'a':1, 'b': 2, 'c':5}
y = {'b':10, 'c': 11}
z = {'e':5, 'b': 2}

我想要一个像这样的合并字典。

f = {'a':[1,0,0], 'b': [2,10,2], 'c':[5,11,0], 'e':[0,0,5]}

有没有迭代的方法来做到这一点。在创建最终字典后，我想创建一个矩阵，其中a，b，c，e将是具有三行的列标题，如下所示：

Answer 1

keys = ('a', 'b', 'c', 'e')
dicts = (x, y, z)
f = {k: [d.get(k, 0) for d in dicts] for k in keys}

或者作为一个单行：

f = {k: [d.get(k, 0) for d in (x, y, z)] for k in ('a', 'b', 'c', 'e')}

Answer 2

您可以使用dict.viewkeys和reduce功能获取密钥，然后使用collections.defaultdict将相应的值附加到密钥：

>>> x = {'a':1, 'b': 2, 'c':5}
>>> y = {'b':10, 'c': 11}
>>> z = {'e':5, 'b': 2}
>>> l=[x,y,z]

>>> keys=reduce(lambda i,j: i | j ,[j.viewkeys() for j in l])
>>> from collections import defaultdict
>>> d=defaultdict(list)
>>> for k in keys :
...    for i in l:
...        d[k].append(i.get(k,0))
... 
>>> d
defaultdict(<type 'list'>, {'a': [1, 0, 0], 'c': [5, 11, 0], 'b': [2, 10, 2], 'e': [0, 0, 5]})

然后你可以创建合适的矩阵：

>>> [[i]+j for i,j in d.items()]
[['a', 1, 0, 0], ['c', 5, 11, 0], ['b', 2, 10, 2], ['e', 0, 0, 5]]

Answer 3

以下是执行这两项任务的方法：

from collections import defaultdict

dicts = [{'a':1, 'b': 2, 'c':5},
         {'b':10, 'c': 11},
         {'e':5, 'b': 2}]

# determine what keys exist in the dictionaries and sort them
keys = sorted(reduce(lambda a,b: a|b, (set(d) for d in dicts)))

f = defaultdict(list)
for k in keys:
    for d in dicts:
        f[k].append(d.get(k, 0))

print 'merged\n  f: {}'.format(dict(f))

matrix = [keys] + [[d.get(k, 0) for k in keys] for d in dicts]
print '\nmatrix:'
for row in matrix:
    print ' '.join('{:>2}'.format(elem) for elem in row)

输出：

merged
  f: {'a': [1, 0, 0], 'c': [5, 11, 0], 'b': [2, 10, 2], 'e': [0, 0, 5]}

matrix:
 a  b  c  e
 1  2  5  0
 0 10 11  0
 0  2  0  5

合并python词典

3 个答案: