Python相当于自定义类的C ++ begin()和end() [英] Python equivalent of C++ begin() and end() for custom classes
问题描述
假设您有一个字符,其键是整数。值也是字典,其键是字符串,其值是numpy数组。
类似于:
Say you have a dictionary whose keys are integers. The values are also dictionaries whose keys are strings and whose values are numpy arrays. Something like:
custom = {1: {'a': np.zeros(10), 'b': np.zeros(100)}, 2:{'c': np.zeros(20), 'd': np.zeros(200)}}
我在代码中一直使用这个自定义数据结构,每次我需要迭代这个结构的numpy数组中的每一行,我必须这样做:
I've been using this custom data structure quite a lot in the code, and every time I need to iterate over each of the rows in the numpy arrays of this structure, I have to do:
for d, delem in custom.items():
for k, v in delem.items():
for row in v:
print(row)
是否可以在函数àla C ++中封装此行为,您可以在其中实际实现自定义 begin()
和端()
?此外,迭代器还应该包含有关其相应字典中的键的信息。我设想类似于:
Is it possible to encapsulate this behavior in functions à la C++ where you can actually implement custom begin()
and end()
? Also, the iterator should also have information about the keys in their corresponding dictionaries. I envision something like:
for it in custom:
d, e, row = *it
# then do something with these
推荐答案
import numpy as np
custom = {
1: {'a': np.zeros(10), 'b': np.zeros(100)},
2:{'c': np.zeros(20), 'd': np.zeros(200)}
}
my_gen = (
(key, subkey, np_array)
for (key, a_dict) in custom.items()
for subkey, np_array in a_dict.items()
)
for key, subkey, np_array in my_gen:
print(key, subkey, np_array)
--output:--
1 b [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
1 a [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
2 d [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0.]
2 c [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0.]
或者,您可以重建数据结构为你的目的更有用的东西:
Or, you could reconstitute your data structure into something that is more useful for your purposes:
import numpy as np
custom = {
1: {'a': np.zeros(10), 'b': np.zeros(100)},
2:{'c': np.zeros(20), 'd': np.zeros(200)}
}
#Create a *list* of tuples:
converted_data = [
(np_array, subkey, key)
for (key, a_dict) in custom.items()
for subkey, np_array in a_dict.items()
]
for np_array, subkey, key in converted_data:
print(key, subkey, np_array)
创建自定义迭代器:
class Dog:
def __init__(self, data):
self.data = data
self.max = len(data)
self.index_pointer = 0
def __next__(self):
index = self.index_pointer
if index < self.max:
current_val = self.data[index]
self.index_pointer += 1
return current_val
else:
raise StopIteration
class MyIter:
def __iter__(self):
return Dog([1, 2, 3])
for i in MyIter():
print(i)
--output:--
1
2
3
__ iter __()
只需要返回一个实现的对象__next __()
方法,所以你可以将这两个类组合起来:
__iter__()
just needs to return an object that implements a __next__()
method, so you can combine those two classes like this:
class MyIter:
def __init__(self, data):
self.data = data
self.max = len(data)
self.index_pointer = 0
def __iter__(self):
return self #I have a __next__() method, so let's return me!
def __next__(self):
index = self.index_pointer
if index < self.max:
current_val = self.data[index]
self.index_pointer += 1
return current_val
else:
raise StopIteration
for i in MyIter([1, 2, 3]):
print(i)
--output:--
1
2
3
更复杂 __ next __()
方法:
import numpy as np
class CustomIter:
def __init__(self, data):
self.data = data
self.count = 0
def __iter__(self):
return self
def __next__(self):
count = self.count
self.count += 1
if count == 0: #On first iteration, retun a sum of the keys
return sum(self.data.keys())
elif count == 1: #On second iteration, return the subkeys in tuples
subkeys = [
a_dict.keys()
for a_dict in self.data.values()
]
return subkeys
elif count == 2: #On third iteration, return the count of np arrays
np_arrays = [
np_array
for a_dict in self.data.values()
for np_array in a_dict.values()
]
return len(np_arrays)
else: #Quit after three iterations
raise StopIteration
custom = {
1: {'a': np.zeros(10), 'b': np.zeros(100)},
2:{'c': np.zeros(20), 'd': np.zeros(200)}
}
for i in CustomIter(custom):
print(i)
--output:--
3
[dict_keys(['b', 'a']), dict_keys(['d', 'c'])]
4
这篇关于Python相当于自定义类的C ++ begin()和end()的文章就介绍到这了,希望我们推荐的答案对大家有所帮助,也希望大家多多支持IT屋!