Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

How to split data into training data and test data?

Reply
New Contributor
Posts: 4

How to split data into training data and test data?

Im using SAS 9.4 

So, I have a car data. I want to split data into training data and test data based on a variable "model year".

I want model year which is even to be training data and model year which is odd to be test data.

I tried the following code:

 

proc import out=work.car5
datafile= "C:\desktop\Auto1.csv" dbms=csv;
getnames=yes; datarow=2;
run;

data train test; 
set year;
b=mod(year,2);
if b=0
then output train;
else output test;
run;

 

but i got some error message. it did not work. How to solve this problem??  Thanks in advance!! 

 

Valued Guide
Posts: 580

Re: How to split data into training data and test data?

Posted in reply to ainlayray

ainlayray wrote:

...

 but i got some error message. it did not work. How to solve this problem??  Thanks in advance!! 

 


This is normal logic: if you get an error message something did not work. So please post the log that we can see that message.

New Contributor
Posts: 4

Re: How to split data into training data and test data?

Posted in reply to andreas_lds

NOTE: The SAS System stopped processing this step because of errors.
WARNING: The data set WORK.TRAIN may be incomplete. When this step was stopped there were 0
observations and 2 variables.
WARNING: The data set WORK.TEST may be incomplete. When this step was stopped there were 0
observations and 2 variables.

 

oh i see. im new here. This is the error message after the code. It created 2 new sets but 0 observations...

Valued Guide
Posts: 580

Re: How to split data into training data and test data?

Posted in reply to ainlayray

Sorry, my request was misleading: please post the complete log including all steps that are executed.

Highlighted
New Contributor
Posts: 4

Re: How to split data into training data and test data?

Posted in reply to andreas_lds

thanks!

okay, I will post in that format in my next posts.  I will post the complete code and log.

 

And I just solved my problem. In my command: "set year" , it refers to year data, not year variable.

so, I changed that to "set car5" which refers to my data. Then it worked... It is like many beginners' mistake...

 

Thanks anyways!!

Ask a Question
Discussion stats
  • 4 replies
  • 240 views
  • 1 like
  • 2 in conversation