您正在关注此问题

您将在关注的内容源中看到更新。
您可能会收到电子邮件，具体取决于您的通信预设项。

how could avoid too have loop in big data

1 次查看（过去 30 天）

显示更早的评论

Nooshin Mahmoodi 2018-10-21

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data

⋮

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data

评论： jonas 2018-10-23

采纳的回答： jonas

I have a big data set(near to 13G). it is a table with information about different agencies at a different time(every day of 10 years). I use unique() to make groups for months and years and agencies. (is it true?) I need to calculate the mean of each agency at all month and years.how could I save time?

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

采纳的回答

jonas 2018-10-21

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#answer_342561

⋮

1
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#answer_342561

编辑：jonas 2018-10-21

You can use findgroups/splitapply or put the data in a table and use varfun or grpstats.

Upload some data if you need code.

8 个评论
显示 6更早的评论隐藏 6更早的评论

Nooshin Mahmoodi 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626233

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626233

test.xls

the point is that some of my data is in Persian and my date too,e.g, 1392 is persain year, and I have different sheet in excel. I will upload a small part of them. if I used func = @mean; Mean = varfun(func,data,'GroupingVariables',شهرستان) I received this error: Unrecognized variable name 'اسكو'.

jonas 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626250

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626250

编辑：jonas 2018-10-22

在 MATLAB Online 中打开

Persian characters are not accepted in variable names, at least on my machine. I can load the characters fine in the table, but reading the table displays a warning that the variable names were changed. Easiest would be to change the of the variables. For a table T, with 17 columns, you can write:

T.Properties.VariableNames = sprintfc('Var_%g',1:17)

Perhaps you already know this, but make sure you only pass numeric values to varfun. You can pass a table with non-numeric data to varfun, but you must then specify the numeric columns that you want to evaluate. For example,

vmean = varfun(@mean,T,'InputVariables','Var_1','GroupingVariables','Var_14')

Var_14 is persian text and it works fine!

Nooshin Mahmoodi 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626275

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626275

THANKS ALOT, but even by changing the name of the variables, I still have error. If I used the unique and cell it works but if I want to get mean cities in different months and years...I need to use loop

jonas 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626277

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626277

编辑：jonas 2018-10-22

I edited my comment a bit after you replied. Consider my edit, try again and then give me the entire error message and explain what columns you want to use as grouping variables (column 2 and... 14?)

I'm having some trouble parsing the dates correctly, they are interpreted in persian for some reason. If you can somehow get the dates in datetime format, then you can easily add two more columns with the month and year and add them as additional grouping variables.

Nooshin Mahmoodi 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626311

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626311

I use this code for months and years:1391 and 1392 are years month = unique(month(datetime(data.(16),'InputFormat','yyyy/MM/dd'))); year=unique(year(datetime(data.(16),'InputFormat','yyyy/MM/dd'))); but I want to group data by the City... but it gives the error like the last time Error using getVarIndices (line 47) Unrecognized variable name 'اسكو'.

Error in table/varfun (line 103) groupVars = getVarIndices(a,groupVars);

Error in index (line 44) Mean = varfun(func,data,'GroupingVariables',City)

jonas 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626322

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626322

编辑：jonas 2018-10-22

在 MATLAB Online 中打开

The grouping variable must be a string. If your variable is called "City" then the syntax is as follows:

varfun(func,data,'GroupingVariables','City')

I suspect that this will return a different error, unless the table data includes only numerical columns, for reasons explained in my previous comment.

If you already have the dates in datetime format, then the last steps are quite straight-forward. Let's say you have a column with your dates called dates, you will want to:

Add two columns to your table with month number and year number

T.Yrs = year(T.dates)
T.Mnth = month(T.dates)

Add these to the grouping variables

varfun(func,data,'GroupingVariables',{'City','Yrs','Mnth');

Add the variable names you want to evaluate, as I described above.

I'm going to sleep, but will check in tomorrow. Also, you may want to try using english variable names, in the event that there is some problem with persian variable names (I have never tested).

Nooshin Mahmoodi 2018-10-22

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626332

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626332

thank a lot for your answer and time

jonas 2018-10-23

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626478

⋮

链接

此评论的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/425242-how-could-avoid-too-have-loop-in-big-data#comment_626478

My pleasure!

请先登录，再进行评论。

另请参阅

类别

MATLAB Language Fundamentals Data Types Dates and Time

在 Help Center 和 File Exchange 中查找有关 Dates and Time 的更多信息

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

发生错误

由于页面发生更改，无法完成操作。请重新加载页面以查看其更新后的状态。

Translated by

选择网站

选择网站以获取翻译的可用内容，以及查看当地活动和优惠。根据您的位置，我们建议您选择：。

(English)
(Deutsch)
(Français)

（简体中文）
(English)

您也可以从以下列表中选择网站：

如何获得最佳网站性能

选择中国网站（中文或英文）以获得最佳网站性能。其他 MathWorks 国家/地区网站并未针对您所在位置的访问进行优化。

美洲

América Latina (Español)
Canada (English)
United States (English)

欧洲

Belgium (English)
Denmark (English)
Deutschland (Deutsch)
España (Español)
Finland (English)
France (Français)
Ireland (English)
Italia (Italiano)
Luxembourg (English)

Netherlands (English)
Norway (English)
Österreich (Deutsch)
Portugal (English)
Sweden (English)
Switzerland
United Kingdom(English)

亚太

Australia (English)
India (English)
New Zealand (English)
中国
- 简体中文Chinese
- English
日本Japanese (日本語)
한국Korean (한국어)

联系您当地的办事处