I’m new to SAS and need to know how to create neural network or decision tree on SAS Enterprise Guide. Do you know a link to document or video that shows the process? Thanks
Hi Arroqi, This type of question falls outside of the scope of the CZcams video and is not the kind of question we can answer here. Your license for SAS needs to be checked to see if you have the correct components for creating Neural Networks and Decision Trees. Typically, you would use SAS Enterprise Miner, not SAS Enterprise Guide for Neural Networks and Decision Trees. Please review this video: 2.sas.com/60543af0o it is the first in a series of 6 videos that shows how to use SAS Enterprise Miner. Other videos in the series show Decision Trees and Neural Networks. However, your best resources for questions like this is to work with SAS Tech Support. To open a track with Tech Support, fill out the form at this link: 2.sas.com/60563af0q .
It's only available in SAS Studio within SAS Viya as a snippet under the "SAS Viya Machine Learning" category. If you don't have access to SAS Viya currently, you can experience it in our trial environment 2.sas.com/6057GQgqV Hope this helps!
Hi Adrija, The instructor for this video used PROC SGPLOT to plot the percentage of missing values However, there were some steps before the plotting step that prepared the data for plotting. We've asked the instructor to elaborate on what the steps were and will post back when we have that information. Currently the instructor is not available but we expect that she will be available next week to comment on this question. In the meantime, if you want to find out whether your variables have missing values, you can investigate PROC FREQ using the NLEVELS option with the MISSING option on the TABLES statement. This is one way to determine whether a variable has missing values and what the percentage of missing values is.
Hi Adrija, The video instructor has explained that she used PROC CARDINALITY (in SAS Viya) to create an output dataset that contained the NMISS information. You'll find these snippets that are included under the SAS Studio - SAS Viya Machine Learning snippets. Specifically, if you have SAS Viya with the Machine Learning snippets, you'd use the Prepare and Explore Data snippet, which is what she was running at that time point in the video. This SAS Studio snippet uses PROC CARDINALITY to get the _NMISS_ information (number of missing observations for that variable). Then, the data set from PROC CARDINALY is sent to PROC SGPLOT to display that _NMISS_ information in graphical form. This approach will only work in SAS Viya installations. However, if you do not have access to PROC CARDINALITY, then you can get the NMISS statistic from PROC MEANS too. And you could plot the information using PROC SGPLOT. This article for UCLA students shows an example of getting NMISS information using PROC MEANS: 2.sas.com/6052MIBwk .
Thanks to SAS management , This video l watched make me to learn more on Data logistics in categories of regression.
had a 'Hold on'!!! moment when I saw you using %if in open code, unfortunately is not available in my version of EG. Great video regardless thanks
Nice tutorial on Logistic Regression!! Thanks.
Thanks for your feedback, we're glad you enjoyed it!
is the data still available the link does not appear to work ?
Which link are you referring to?
I’m new to SAS and need to know how to create neural network or decision tree on SAS Enterprise Guide. Do you know a link to document or video that shows the process? Thanks
Thank you for your inquiry! We will check in to this for you!
Hi Arroqi,
This type of question falls outside of the scope of the CZcams video and is not the kind of question we can answer here. Your license for SAS needs to be checked to see if you have the correct components for creating Neural Networks and Decision Trees. Typically, you would use SAS Enterprise Miner, not SAS Enterprise Guide for Neural Networks and Decision Trees. Please review this video: 2.sas.com/60543af0o it is the first in a series of 6 videos that shows how to use SAS Enterprise Miner. Other videos in the series show Decision Trees and Neural Networks. However, your best resources for questions like this is to work with SAS Tech Support. To open a track with Tech Support, fill out the form at this link: 2.sas.com/60563af0q .
Dear Crista: How do I download the SAS Codes which you used in the logistic regression? Thanks. Professor Dey
It's only available in SAS Studio within SAS Viya as a snippet under the "SAS Viya Machine Learning" category. If you don't have access to SAS Viya currently, you can experience it in our trial environment 2.sas.com/6057GQgqV
Hope this helps!
Hi! By running which code you got the percentage of missing values? It was around 7:20 minutes into the video.
Adrija, thank you for your inquiry! We are checking on this for you!
Hi Adrija,
The instructor for this video used PROC SGPLOT to plot the percentage of missing values However, there were some steps before the plotting step that prepared the data for plotting. We've asked the instructor to elaborate on what the steps were and will post back when we have that information. Currently the instructor is not available but we expect that she will be available next week to comment on this question.
In the meantime, if you want to find out whether your variables have missing values, you can investigate PROC FREQ using the NLEVELS option with the MISSING option on the TABLES statement. This is one way to determine whether a variable has missing values and what the percentage of missing values is.
Thank you so much for the information. Looking forward to knowing the whole code.
You are most welcome!
Hi Adrija,
The video instructor has explained that she used PROC CARDINALITY (in SAS Viya) to create an output dataset that contained the NMISS information. You'll find these snippets that are included under the SAS Studio - SAS Viya Machine Learning snippets. Specifically, if you have SAS Viya with the Machine Learning snippets, you'd use the Prepare and Explore Data snippet, which is what she was running at that time point in the video. This SAS Studio snippet uses PROC CARDINALITY to get the _NMISS_ information (number of missing observations for that variable). Then, the data set from PROC CARDINALY is sent to PROC SGPLOT to display that _NMISS_ information in graphical form.
This approach will only work in SAS Viya installations. However, if you do not have access to PROC CARDINALITY, then you can get the NMISS statistic from PROC MEANS too. And you could plot the information using PROC SGPLOT. This article for UCLA students shows an example of getting NMISS information using PROC MEANS: 2.sas.com/6052MIBwk .