Python：迭代两个以list为值的字典

Question

我构建了两个包含图像直方图值的字典。每个字典都以图像文件的文件名为键，将三个一维向量的列表放在一起作为其值。在

示例：{'someFileName.jpg' : ['forecolor=2,3,5,5,6','edge=2,4,5','texture=5,4,3']}

以下是我的一本词典的实际表现：

字典1

{'/Users/images/Transcend-8GB-Class-10-SDHC-Flash-Memory-Card.jpg': ['fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3', 'edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63', 'texture=1,78,27,37,13,6,6,7,78']}

词典2

^{pr2}$

我的最终目标是将两个字典传递给一个方法并实际运行cosign值

例如：每个字典都有list作为它的值，所以对于每个dictionary key，我要在dictionary1的key1，velu1和dictionary2的key1，value1进行向量乘法

我有向量乘法函数，所以我想知道如何正确地迭代，我想用一个屈服函数，但当我尝试的时候，它并没有真正起作用。到目前为止，我得到的是：

def cosignSimilarity(image1VectorDict, image2VectorDict):
    for image1Key, image2Value in image1VectorDict.iteritems():
        print image1Key
        for aValue in image1Value:
            print aValue
            for image2Key, image2Value in image2VectorDict.iteritems():
                for eValue in image2Value:
                    print aValue
                    print "\n"
                    print eValue

仅供参考：我不是在求余弦计算的帮助。在

这就是我当前的代码输出数据的方式如果我能将键与键从一个字典分离到另一个字典，那么我就可以完成剩下的工作，比如计算余弦值。在

   First  Dictionary
    {'/Users/test/Transcend-8GB-Class-10-SDHC-Flash-Memory-Card.jpg': ['fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3', 'edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63', 'texture=1,78,27,37,13,6,6,7,78']}
    ------------------
Second Dictionary
    {'/Users/test/kodax-camera-M531.jpg': ['fcolor=2,74,6,20,30,1,2,0,1,0,0,0,1,3,2,0,0,0,0,0,1,1,1,0,0,2,0,0,0,2,2,0,0,0,0,0,2,2,1,0,0,5,0,0,0,1,4,0,0,0,0,0,2,2,1,0,0,1,0,0,0,3,1,0,0,0,0,0,1,1,0,0,0,3,0,0,0,1,2,0,0,0,0,0,2,2,1,0,0,4,0,0,0,0,5,0,0,0,0,0,2,1,0,0,0,0,0,0,0,3,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0', 'edge=1,4,1,88,128,22,8,39,25,142,230,226,31,60,64,255,252,12,76', 'texture=1,15,32,31,28,19,16,12,98']}
    ++++++++++++++++++
    /Users/test/Transcend-8GB-Class-10-SDHC-Flash-Memory-Card.jpg
    fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3
    fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3


    fcolor=2,74,6,20,30,1,2,0,1,0,0,0,1,3,2,0,0,0,0,0,1,1,1,0,0,2,0,0,0,2,2,0,0,0,0,0,2,2,1,0,0,5,0,0,0,1,4,0,0,0,0,0,2,2,1,0,0,1,0,0,0,3,1,0,0,0,0,0,1,1,0,0,0,3,0,0,0,1,2,0,0,0,0,0,2,2,1,0,0,4,0,0,0,0,5,0,0,0,0,0,2,1,0,0,0,0,0,0,0,3,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0
    fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3


    edge=1,4,1,88,128,22,8,39,25,142,230,226,31,60,64,255,252,12,76
    fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3


    texture=1,15,32,31,28,19,16,12,98
    edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63
    edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63


    fcolor=2,74,6,20,30,1,2,0,1,0,0,0,1,3,2,0,0,0,0,0,1,1,1,0,0,2,0,0,0,2,2,0,0,0,0,0,2,2,1,0,0,5,0,0,0,1,4,0,0,0,0,0,2,2,1,0,0,1,0,0,0,3,1,0,0,0,0,0,1,1,0,0,0,3,0,0,0,1,2,0,0,0,0,0,2,2,1,0,0,4,0,0,0,0,5,0,0,0,0,0,2,1,0,0,0,0,0,0,0,3,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0
    edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63


    edge=1,4,1,88,128,22,8,39,25,142,230,226,31,60,64,255,252,12,76
    edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63


    texture=1,15,32,31,28,19,16,12,98
    texture=1,78,27,37,13,6,6,7,78
    texture=1,78,27,37,13,6,6,7,78


    fcolor=2,74,6,20,30,1,2,0,1,0,0,0,1,3,2,0,0,0,0,0,1,1,1,0,0,2,0,0,0,2,2,0,0,0,0,0,2,2,1,0,0,5,0,0,0,1,4,0,0,0,0,0,2,2,1,0,0,1,0,0,0,3,1,0,0,0,0,0,1,1,0,0,0,3,0,0,0,1,2,0,0,0,0,0,2,2,1,0,0,4,0,0,0,0,5,0,0,0,0,0,2,1,0,0,0,0,0,0,0,3,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0
    texture=1,78,27,37,13,6,6,7,78


    edge=1,4,1,88,128,22,8,39,25,142,230,226,31,60,64,255,252,12,76
    texture=1,78,27,37,13,6,6,7,78


    texture=1,15,32,31,28,19,16,12,98

很明显，正如你所看到的，我正在吐出许多相同价值的重复

以下是我正在处理的实际字典：

字典1:

{'/Users/test/Transcend-8GB-Class-10-SDHC-Flash-Memory-Card.jpg': ['fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3', 'edge=1,252,1,32,124,194,63,252,67,15,240,1,7,244,66,47,0,192,63', 'texture=1,78,27,37,13,6,6,7,78']}

字典2:

{'/Users/test/kodax-camera-M531.jpg': ['fcolor=2,74,6,20,30,1,2,0,1,0,0,0,1,3,2,0,0,0,0,0,1,1,1,0,0,2,0,0,0,2,2,0,0,0,0,0,2,2,1,0,0,5,0,0,0,1,4,0,0,0,0,0,2,2,1,0,0,1,0,0,0,3,1,0,0,0,0,0,1,1,0,0,0,3,0,0,0,1,2,0,0,0,0,0,2,2,1,0,0,4,0,0,0,0,5,0,0,0,0,0,2,1,0,0,0,0,0,0,0,3,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0', 'edge=1,4,1,88,128,22,8,39,25,142,230,226,31,60,64,255,252,12,76', 'texture=1,15,32,31,28,19,16,12,98']}

我有lamba函数

cosinLamba = lambda a, b : round(NP.inner(a, b)/(LA.norm(a)*LA.norm(b)), 3)

我想遍历dictionary 1和dictionary 2，得到dictionary1的fcolor值'fcolor=2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3'

和字典2的fcolor值

'fcolor=2,74,6,20,30,1,2,0,1,0,0,0,1,3,2,0,0,0,0,0,1,1,1,0,0,2,0,0,0,2,2,0,0,0,0,0,2,2,1,0,0,5,0,0,0,1,4,0,0,0,0,0,2,2,1,0,0,1,0,0,0,3,1,0,0,0,0,0,1,1,0,0,0,3,0,0,0,1,2,0,0,0,0,0,2,2,1,0,0,4,0,0,0,0,5,0,0,0,0,0,2,1,0,0,0,0,0,0,0,3,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0'

把它们发送给我的lamba函数cosinLamba(valu1, value2)value1和value2是字符串，这就是为什么我把它们作为值保存在字典中的原因。我想对fcolor，texture，edge，所有我在每个字典中为给定图片存储的向量。在

Answer 1

我不熟悉您要执行的余弦计算，但除此之外，如果我对您的问题的理解是正确的，那么以下代码应该可以工作：

for key1, vals1 in dict1.iteritems():
    vals2 = dict2[key1]
    for val1, val2 in zip(vals1, vals2):
        # you now have the corresponding values for each image file
        compute_cosign(val1, val2)

Answer 2

您可以先将表示更改为：

{'someFileName.jpg' : {'forecolor': [2,3,5,5,6],'edge': [2,4,5],'texture':[5,4,3]}}

或者

^{pr2}$

例如，要获取第一个案例的对应列表：

from itertools import product

# pair info for each image with info of every image from another dictionary
for (fn1, d1), (fn2,d2) in product(dict1.iteritems(), dict2.iteritems()):
    for property_, list_value in d1.iteritems():
        compute_cosine_similarity(list_value, d2[property_])

使用字符串列表的表示形式，如下所示：

from itertools import product

# pair info for each image with info of every image from another dictionary
for (fn1,lst1), (fn2,lst2) in product(dict1.iteritems(), dict2.iteritems()):
    # assume all lists has the same order of elements
    for string_value1, string_value2 in zip(lst1, lst2):
        compute(string_value1, string_value2)

不应该将数字存储为ascii字符串列表。如果需要节省内存，可以使用numpy数组。cosinLamba已经接受了它们。在

from collections import namedtuple
import numpy as np

Info = namedtuple('Info', 'forecolor edge texture')

dict1 = {'someFileName.jpg': Info(np.array([...], dtype=np.uint8),
                                  np.array([...], dtype=np.uint8),
                                  np.array([...], dtype=np.uint8))}

调用cosine_similarity（）的代码与您的表示完全相同。在