- android - 多次调用 OnPrimaryClipChangedListener
- android - 无法更新 RecyclerView 中的 TextView 字段
- android.database.CursorIndexOutOfBoundsException : Index 0 requested, 光标大小为 0
- android - 使用 AppCompat 时,我们是否需要明确指定其 UI 组件(Spinner、EditText)颜色
有时,似乎我使用Python(和Pandas)越多,我理解的就越少。所以,如果我只是只见树木不见森林,我深表歉意,但我一直在原地打转,只是看不出自己做错了什么。
基本上,我有一个示例脚本(我想在更大的数据帧上实现),但我无法让它达到我满意的效果。
数据框由各种数据类型的列组成。我想将数据框分组为两列,然后生成一个新的数据框,其中包含每组中每个变量的所有唯一值的列表。 (最终,我想将列表项连接成一个字符串 - 但这是一个不同的问题。)
我使用的初始脚本是:
import numpy as np
import pandas as pd
def tempFuncAgg(tempVar):
tempList = set(tempVar.dropna()) # Drop NaNs and create set of unique values
print(tempList)
return tempList
# Define dataframe
tempDF = pd.DataFrame({ 'id': [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20],
'date': ["02/04/2015 02:34","06/04/2015 12:34","09/04/2015 23:03","12/04/2015 01:00","15/04/2015 07:12","21/04/2015 12:59","29/04/2015 17:33","04/05/2015 10:44","06/05/2015 11:12","10/05/2015 08:52","12/05/2015 14:19","19/05/2015 19:22","27/05/2015 22:31","01/06/2015 11:09","04/06/2015 12:57","10/06/2015 04:00","15/06/2015 03:23","19/06/2015 05:37","23/06/2015 13:41","27/06/2015 15:43"],
'gender': ["male","female","female","male","male","female","female",np.nan,"male","male","female","male","female","female","male","female","male","female",np.nan,"male"],
'age': ["young","old","old","old","old","old",np.nan,"old","old","young","young","old","young","young","old",np.nan,"old","young",np.nan,np.nan]})
# Groupby based on 2 categorical variables
tempGroupby = tempDF.groupby(['gender','age'])
# Aggregate for each variable in each group using function defined above
dfAgg = tempGroupby.agg(lambda x: tempFuncAgg(x))
print(dfAgg)
此脚本的输出符合预期:一系列包含值集的行和包含返回集的数据帧:
{'09/04/2015 23:03', '21/04/2015 12:59', '06/04/2015 12:34'}
{'01/06/2015 11:09', '12/05/2015 14:19', '27/05/2015 22:31', '19/06/2015 05:37'}
{'15/04/2015 07:12', '19/05/2015 19:22', '06/05/2015 11:12', '04/06/2015 12:57', '15/06/2015 03:23', '12/04/2015 01:00'}
{'02/04/2015 02:34', '10/05/2015 08:52'}
{2, 3, 6}
{18, 11, 13, 14}
{4, 5, 9, 12, 15, 17}
{1, 10}
date \
gender age
female old set([09/04/2015 23:03, 21/04/2015 12:59, 06/04...
young set([01/06/2015 11:09, 12/05/2015 14:19, 27/05...
male old set([15/04/2015 07:12, 19/05/2015 19:22, 06/05...
young set([02/04/2015 02:34, 10/05/2015 08:52])
id
gender age
female old set([2, 3, 6])
young set([18, 11, 13, 14])
male old set([4, 5, 9, 12, 15, 17])
young set([1, 10])
当我尝试将集合转换为列表时出现问题。奇怪的是,它生成 2 个包含相同列表的重复行,但随后失败并出现“ValueError:函数不会减少”错误。
def tempFuncAgg(tempVar):
tempList = list(set(tempVar.dropna())) # This is the only difference
print(tempList)
return tempList
tempDF = pd.DataFrame({ 'id': [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20],
'date': ["02/04/2015 02:34","06/04/2015 12:34","09/04/2015 23:03","12/04/2015 01:00","15/04/2015 07:12","21/04/2015 12:59","29/04/2015 17:33","04/05/2015 10:44","06/05/2015 11:12","10/05/2015 08:52","12/05/2015 14:19","19/05/2015 19:22","27/05/2015 22:31","01/06/2015 11:09","04/06/2015 12:57","10/06/2015 04:00","15/06/2015 03:23","19/06/2015 05:37","23/06/2015 13:41","27/06/2015 15:43"],
'gender': ["male","female","female","male","male","female","female",np.nan,"male","male","female","male","female","female","male","female","male","female",np.nan,"male"],
'age': ["young","old","old","old","old","old",np.nan,"old","old","young","young","old","young","young","old",np.nan,"old","young",np.nan,np.nan]})
tempGroupby = tempDF.groupby(['gender','age'])
dfAgg = tempGroupby.agg(lambda x: tempFuncAgg(x))
print(dfAgg)
但是现在的输出是:
['09/04/2015 23:03', '21/04/2015 12:59', '06/04/2015 12:34']
['09/04/2015 23:03', '21/04/2015 12:59', '06/04/2015 12:34']
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
...
ValueError: Function does not reduce
任何解决此问题的帮助将不胜感激,如果我没有看到明显的情况,我提前表示歉意。
编辑顺便说一句,将集合转换为元组而不是列表是没有问题的。
最佳答案
pandas 中的列表有时会出现奇怪的问题。您可以:
使用元组(正如您已经注意到的)
如果您确实需要列表,只需在第二个操作中执行即可,如下所示:
dfAgg.applymap(lambda x: list(x))
完整示例:
import numpy as np
import pandas as pd
def tempFuncAgg(tempVar):
tempList = set(tempVar.dropna()) # Drop NaNs and create set of unique values
print(tempList)
return tempList
# Define dataframe
tempDF = pd.DataFrame({ 'id': [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20],
'date': ["02/04/2015 02:34","06/04/2015 12:34","09/04/2015 23:03","12/04/2015 01:00","15/04/2015 07:12","21/04/2015 12:59","29/04/2015 17:33","04/05/2015 10:44","06/05/2015 11:12","10/05/2015 08:52","12/05/2015 14:19","19/05/2015 19:22","27/05/2015 22:31","01/06/2015 11:09","04/06/2015 12:57","10/06/2015 04:00","15/06/2015 03:23","19/06/2015 05:37","23/06/2015 13:41","27/06/2015 15:43"],
'gender': ["male","female","female","male","male","female","female",np.nan,"male","male","female","male","female","female","male","female","male","female",np.nan,"male"],
'age': ["young","old","old","old","old","old",np.nan,"old","old","young","young","old","young","young","old",np.nan,"old","young",np.nan,np.nan]})
# Groupby based on 2 categorical variables
tempGroupby = tempDF.groupby(['gender','age'])
# Aggregate for each variable in each group using function defined above
dfAgg = tempGroupby.agg(lambda x: tempFuncAgg(x))
# Transform in list
dfAgg.applymap(lambda x: list(x))
print(dfAgg)
pandas 中有很多这样的奇怪行为,通常最好继续使用解决方法(像这样),而不是寻找完美的解决方案
关于python - 使用 Pandas groupby agg 函数将集合转换为列表会导致 'ValueError: Function does not reduce',我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/36322217/
如标题所示,ans_list是一个答案列表,ans_index是一个数字(答案在词汇表中的索引,但与atm无关) 这里生成的 tree.anslist 是什么? (例如,仅针对第一个),忽略迭代。 f
我目前将用户的输入存储在逗号分隔的列表中,如下所示: Userid | Options 1 | 1,2,5 用户在一个数组形式中勾选一组选项,然后用逗号连接起来 1,2,5 然后 MySQ
我目前将用户的输入存储在逗号分隔的列表中,如下所示: Userid | Options 1 | 1,2,5 用户在一个数组形式中勾选一组选项,然后用逗号连接起来 1,2,5 然后 MySQ
我想知道如何完全展平列表和包含它们的东西。除其他外,我想出了一个解决方案,它可以将具有多个元素的东西滑倒并将它们放回原处,或者在滑倒后将具有一个元素的东西拿走。 这与 How do I “flatte
我想知道如何完全展平列表和包含它们的东西。除其他外,我想出了一个解决方案,它可以将具有多个元素的东西滑倒并将它们放回原处,或者在滑倒后将带有一个元素的东西拿走。 这与 How do I “flatte
这个问题已经有答案了: Convert nested list to 2d array (3 个回答) 已关闭 7 年前。 java中有没有快捷方式可以转换 List> 到 String[][] ?
我在排序时遇到问题 List> 。我创建了一个自定义比较器,在其中编写了对数据进行排序的代码。 public class CustomComparator implements Comparator
这个问题已经有答案了: 已关闭10 年前。 Possible Duplicate: Java Generics: Cannot cast List to List? 我只是想知道为什么下面的java代
试图想出一个 LINQy 方法来做到这一点,但我什么也没想到。 我有一个对象列表<>,其中包含一个属性,该属性是逗号分隔的字母代码列表: lst[0].codes = "AA,BB,DD" lst[1
假设我有这些任务: points = [] point = (1, 2) 我怎么会这样做: points += point 它工作得很好,并且给了我点 = [1, 2]。但是,如果我这样做: poin
如何在 scala 中将 List[Task[List[Header]]] 类型转换为 Task[List[Header]]。 我有一个方法返回 Task[List[Header]] 并多次调用 do
如何在 Java 中查找二维列表的元素? 我有一个参数为 List> 的函数我想知道如何找到这个列表的行和列。 最佳答案 如果你喜欢 List> obj 然后你就可以像这样访问 obj.get(cur
分配 List到 List工作正常。 分配 List>到 List>不编译。 代码 public class Main { public static void main(String[] a
我正在用 Java 编写一个方法,该方法必须接收并迭代 Serializable 的 List。 有什么区别: public void myMethod(List list) { } 和 public
我看到很多人想用 mvvm 更新网格/列表/树的一部分,但他们不想刷新整个列表。 对于所有遇到此问题的人,我做了以下示例。 希望这对你有用。 最佳答案 这是一个简单的例子。整个代码中最重要的是: Bi
我正在为现有的 C++ 库编写包装器,该库使用列表,其中 T 是自定义结构。我被建议使用 vector 而不是列表,但我试图避免修改库。 为了更好地理解这个场景,我做了一个简单的应用程序,使用一个列表
List list List list 这两种声明有什么区别吗? 谢谢, 最佳答案 是的。 List可以包含所有派生自 Base 的不同事物的混合物. List包含同质项(从某种意义上说,它们必须全部
有人可以尽可能详细地解释以下类型之间的区别吗? List List List 让我更具体一点。我什么时候想使用 // 1 public void CanYouGiveMeAnAnswer(List l
我有一个元组列表,每个元组都是一对列表。所以我的数据看起来像: mylist = [(['foo', 'bar'], ['bar', 'bar']),(['bar', 'bar'],['bar', '
也许是一个时髦的标题,但我遇到了以下问题: 给定一个类型为 (a * b) list 的列表,我想创建一个类型为 (a * b list) list 的新列表。一个例子: 给定列表 let testL
我是一名优秀的程序员,十分优秀!