- 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/
我正在尝试使用 Spark 从 Cassandra 读取数据。 DataFrame rdf = sqlContext.read().option("keyspace", "readypulse
这是代码: void i_log_ (int error, const char * file, int line, const char * fmt, ...) { /* Get erro
我必须调试一个严重依赖 Gtk 的程序。问题是由于某些原因,在使用 GtkWindow 对象时开始出现许多运行时警告。问题是,即使 Gtk 提示严重错误,它也不会因这些错误而中止。我没有代码库的更改历
我正在尝试从已有效编译和链接的程序中检索二进制文件。我已经通过 GL_PROGRAM_BINARY_LENGTH 收到了它的长度。该文档说有两个实例可能会发生 GL_INVALID_OPERATION
我有一个托管在 Azure 环境中的服务。我正在使用控制台应用程序使用该服务。这样做时,我得到了异常: "The requested service, 'http://xxxx-d.yyyy.be/S
我有以下代码,它被 SEGV 信号杀死。使用调试器表明它被 main() 中的第一个 sem_init() 杀死。如果我注释掉第一个 sem_init() ,第二个会导致同样的问题。我试图弄清楚是什么
目前我正在编写一个应用程序(目标 iOS 6,启用 ARC),它使用 JSON 进行数据传输,使用核心数据进行持久存储。 JSON 数据由 PHP 脚本通过 json_encode 从 MySQL 数
我对 Xamarin.Forms 还是很陌生。我在出现的主页上有一个非常简单的功能 async public Task BaseAppearing() { if (UserID
这是我的代码的简化版本。 public class MainActivity extends ActionBarActivity { private ArrayList entry = new Arr
我想弄明白为什么我的两个 Java 库很难很好地协同工作。这是场景: 库 1 有一个类 A,其构造函数如下: public A(Object obj) { /* boilerplate */ } 在以
如果网站不需要身份验证,我的代码可以正常工作,如果需要,则在打印“已创建凭据”后会立即出现 EXC_BAD_ACCESS 错误。我不会发布任何内容,并且此代码是直接从文档中复制的 - 知道出了什么问题
我在使用 NSArray 填充 UITableView 时遇到问题。我确信我正在做一些愚蠢的事情,但我无法弄清楚。当我尝试进行简单的计数时,我得到了 EXC_BAD_ACCESS,我知道这是因为我试图
我在 UITableViewCell 上有一个 UITextField,在另一个单元格上有一个按钮。 我单击 UITextField(出现键盘)。 UITextField 调用了以下方法: - (BO
我有一个应用程序出现间歇性崩溃。崩溃日志显示了一个堆栈跟踪,这对我来说很难破译,因此希望其他人看到了这一点并能为我指出正确的方向。 基本上,应用程序在启动时执行反向地理编码请求,以在标签中显示用户的位
我开发了一个 CGImage,当程序使用以下命令将其显示在屏幕上时它工作正常: [output_view.layer performSelectorOnMainThread:@selector(set
我正在使用新的 EncryptedSharedPreferences以谷歌推荐的方式上课: private fun securePrefs(context: Context): SharedPrefe
我有一个中继器,里面有一些控件,其中一个是文本框。我正在尝试使用 jquery 获取文本框,我的代码如下所示: $("#").click(function (event) {}); 但我总是得到 nu
在以下场景中观察到 TTS 初始化错误,太随机了。 已安装 TTS 引擎,存在语音集,并且可以从辅助功能选项中播放示例 tts。 TTS 初始化在之前初始化和播放的同一设备上随机失败。 在不同的设备(
maven pom.xml org.openjdk.jol jol-core 0.10 Java 类: public class MyObjectData { pr
在不担心冲突的情况下,可以使用 MD5 作为哈希值,字符串长度最多为多少? 这可能是通过为特定字符集中的每个可能的字符串生成 MD5 哈希来计算的,长度不断增加,直到哈希第二次出现(冲突)。没有冲突的
我是一名优秀的程序员,十分优秀!