嵌套字典分配错误 - 可变字典混淆

Question

我很难在嵌套字典中分配作业。我有一本包含两所学校、三名教师和四名学生的嵌套字典。这只是一个玩具示例，所以我不在乎每所学校的学生和老师都相同。但是，我确实希望每个等级都不同。但是，使用下面的代码，所有班级和所有学校的所有学生的成绩都与最后进入的学校/班级相同。

schools = ['School A', 'School B']
teachers = ['mr. smith', 'ms. jones', 'mr. kronk']
students = ['Adam', 'Nick', 'Jeff', 'Dave']
grade_dict ={}
for i in students:
    grade_dict[i] = ''
    for j in teachers:
        teachers_dict[j] = grade_dict
        for k in schools:
            school_dict[k] = teachers_dict
for k in schools:
    for j in teachers:
        for i in students:
            a = [randint(70, 100), randint(70, 100), randint(70, 100)]
            school_dict[k][j][i] = a

这就是我希望数据的样子：

School A mr. smith Adam [71, 72, 82]
School A mr. smith Nick [86, 80, 96]
School A mr. smith Jeff [77, 70, 83]
School A mr. smith Dave [79, 83, 98]
School A ms. jones Adam [70, 98, 87]
School A ms. jones Nick [80, 94, 76]
School A ms. jones Jeff [79, 82, 93]
School A ms. jones Dave [90, 97, 85]
School A mr. kronk Adam [93, 75, 95]
School A mr. kronk Nick [80, 82, 72]
School A mr. kronk Jeff [75, 72, 89]
School A mr. kronk Dave [86, 92, 98]
School B mr. smith Adam [89, 77, 84]
School B mr. smith Nick [93, 71, 74]
School B mr. smith Jeff [78, 83, 83]
School B mr. smith Dave [72, 83, 70]
School B ms. jones Adam [82, 100, 78]
School B ms. jones Nick [80, 89, 100]
School B ms. jones Jeff [91, 81, 77]
School B ms. jones Dave [86, 86, 74]
School B mr. kronk Adam [82, 73, 100]
School B mr. kronk Nick [81, 71, 74]
School B mr. kronk Jeff [92, 100, 90]
School B mr. kronk Dave [86, 97, 85]

不幸的是，这就是对应字典的样子：

{'School A': {'mr. kronk': {'Adam': [86, 89, 94],
   'Dave': [74, 85, 86],
   'Jeff': [79, 94, 70],
   'Nick': [90, 80, 97]},
  'mr. smith': {'Adam': [86, 89, 94],
   'Dave': [74, 85, 86],
   'Jeff': [79, 94, 70],
   'Nick': [90, 80, 97]},
  'ms. jones': {'Adam': [86, 89, 94],
   'Dave': [74, 85, 86],
   'Jeff': [79, 94, 70],
   'Nick': [90, 80, 97]}},
 'School B': {'mr. kronk': {'Adam': [86, 89, 94],
   'Dave': [74, 85, 86],
   'Jeff': [79, 94, 70],
   'Nick': [90, 80, 97]},
  'mr. smith': {'Adam': [86, 89, 94],
   'Dave': [74, 85, 86],
   'Jeff': [79, 94, 70],
   'Nick': [90, 80, 97]},
  'ms. jones': {'Adam': [86, 89, 94],
   'Dave': [74, 85, 86],
   'Jeff': [79, 94, 70],
   'Nick': [90, 80, 97]}}}

stackoverflow 上似乎有类似的问题，但它们并不能完全帮助我解决这里发生的事情。谢谢。

Answer 1

您可以使用字典理解：

from random import randint
schools = ['School A', 'School B']
teachers = ['mr. smith', 'ms. jones', 'mr. kronk']
students = ['Adam', 'Nick', 'Jeff', 'Dave']

final_dict = {school:{teacher:{student:[randint(70, 100), randint(70, 100), randint(70, 100)] for student in students} for teacher in teachers} for school in schools}

输出：

{'School A': {'mr. smith': {'Nick': [75, 81, 86], 'Dave': [85, 88, 84], 'Adam': [78, 95, 99], 'Jeff': [74, 95, 81]}, 'ms. jones': {'Nick': [76, 86, 92], 'Dave': [92, 100, 95], 'Adam': [98, 99, 90], 'Jeff': [74, 100, 95]}, 'mr. kronk': {'Nick': [84, 97, 79], 'Dave': [93, 91, 89], 'Adam': [83, 98, 79], 'Jeff': [89, 83, 99]}}, 'School B': {'mr. smith': {'Nick': [70, 78, 89], 'Dave': [81, 95, 92], 'Adam': [95, 100, 91], 'Jeff': [91, 83, 82]}, 'ms. jones': {'Nick': [94, 85, 75], 'Dave': [99, 77, 94], 'Adam': [79, 97, 92], 'Jeff': [91, 84, 79]}, 'mr. kronk': {'Nick': [81, 90, 86], 'Dave': [72, 95, 82], 'Adam': [80, 73, 77], 'Jeff': [88, 88, 95]}}}

Answer 2

您在整个字典中（重新）使用相同的 grade_dict 对象。 在每次迭代中创建单独的对象。此外，您不需要在每个键处初始化值：

for i in students:
    for j in teachers:
        teachers_dict[j] = {}
        for k in schools:
            school_dict[k] = teachers_dict

OTOH，您应该通过使用嵌套的 collections.defaultdict 声明 school_dict 来避免使用 for 预分配值/字典。

from collections import defaultdict

school_dict = defaultdict(lambda: school_dict) # dict nesting is recursive

for k in schools:
    for j in teachers:
        for i in students:
            school_dict[k][j][i] = [randint(70, 100), randint(70, 100), randint(70, 100)]