Input:
A= [{'A':0.4, 'B':0.8, 'C':1.0},
{'A':0.5, 'B':0.9, 'C':0.0}]
Desired Output:
A=[{'C':1.0},{'A':0.5, 'B':0.9}]
Note: I wanna iterate through and keep only the key with the maximum value among same keys !! And list of dictionaries should be kept !
A more concise approach based on @Ioannis' answer
max_dict = {k: max([d[k] for d in A]) for k in A[0]}
result = [{k: v for k, v in i.items() if v == max_dict[k]} for i in A]
print(result)
output:
[{'C': 1.0}, {'A': 0.5, 'B': 0.9}]
Detailed approach:
You can create a dictionary with max values and check with it for each list item. You can implement it using 2 for loops.
A = [{'A': 0.4, 'B': 0.8, 'C': 1.0},
{'A': 0.5, 'B': 0.9, 'C': 0.0}]
max_dict = {}
for a_dict in A:
for key, value in a_dict.items():
if key not in max_dict or max_dict[key] < value:
max_dict[key] = value
result = []
for a_dict in A:
temp_dict = {}
for key, value in a_dict.items():
if value == max_dict[key]:
temp_dict[key] = value
result.append(temp_dict)
print(result)
output:
[{'C': 1.0}, {'A': 0.5, 'B': 0.9}]
Ioannis's answer should work, but here is a more visual answer if that didn't make sense.
B={}
for i in range(len(A)):
d = A[i]
for key in d:
if (key not in B): B[key]=i
Now we store the index of each key max in B.
for i in range(len(A)):
d=A[i]
for key in d:
if (A[B[key]][key]<d[key]):
# A[B[key]][key] gets the index of the max using B[key]
# then gets the dictionary at that index using A[B[key]], and then
# then finally gets the value using A[B[key]][key]
B[key]=i
Now, we have all the max indexes. So, we can go through a new list, say C, and add dicts for the length of A (ie there are the same number of dicts as A)
C=[]
for i in range(len(A)):
C.append({})
for key in B:
ind = B[key]
C[ind][key]=A[B[key]][key]
Output (tested on 3.9.4): [{'C': 1.0}, {'A': 0.5, 'B': 0.9}]
B = {k: max([d[k] for d in A]) for k in A[0]}
If B needs to be within a list, then [B].
