Xamius32 Tracker

Re: Is there a way to only use one variable as a two-factor interaction variable

Xamius32 — Wed, 04 Jan 2017 21:59:44 GMT

EM, mostly the regression node. Edited to clarify

Is there a way to only use one variable as a two-factor interaction variable

Xamius32 — Wed, 04 Jan 2017 21:58:48 GMT

i dont want to use all class variables for interactions, I want one specific class variable to be used as an interaction variable with every other input variable.

I am talking mostly about enterprise Miner. For example, in the regression node, under the two-factor interactions option.

Is there a simple way to set this up?

Re: Is there any way to export HP models scoring code?

Xamius32 — Thu, 22 Dec 2016 17:47:28 GMT

I am not sure what the SAS Scoring Accelerator is, but I need a way to see the HP model in SAS code.

i was reading this and I cant seem to find the code. Is there any way to see a coded version of the HP forest model?

https://communities.sas.com/t5/SAS-Communities-Library/Tip-How-Can-I-Apply-HP-Forest-Score-Code-to-Distributed-Data/ta-p/221493

Which says

4. Find a file named HPDMSCORECODE.sas located in the folder corresponding to the Score node in the EMWS workspace directory for the diagram containing your flow. The path to this folder is in the following form:

SAS-Server-Directory/Project-Name/Workspaces/EMWSx/Score-node-ID/

5. Copy the contents of the HPDMSCORECODE.sas file into the SAS Enterprise Miner Program Editor, and run the code. The code from this file is in the correct form to score data in the database once the above librefs and macro variables have been defined.

Is there any way to export HP models scoring code?

Xamius32 — Thu, 22 Dec 2016 16:49:52 GMT

Looking at the SAS Enterprise Miner Tools Production of Score Code I see

that the score code export node is not supported with the HP models like HP forest. I want to make sure that means there is no possible way to get be able to score a HP model in other programs besides SAS.

Thanks

Re: Custom model comparison criteria

Xamius32 — Wed, 21 Dec 2016 16:54:10 GMT

From what I understand the ensemble node can take the average or maximum of different models from the same training data, correct? My goal is a bit different since I am trying to take the maximum of different sets of training data

Custom model comparison criteria

Xamius32 — Wed, 21 Dec 2016 00:12:33 GMT

I am trying to build a unique type of model scoring/comparison. Right now I am testing different models manually and I would like to automatically select which model performs best according to custom criteria.

Basically, I am building a dynamic pricing model, where 25% of users get one of A, B, C, or D prices. I then build one model for each price. So a model is run for everyone who got price A, and B, and so forth.

So then I have 4 models.

I then score all 4 models on my validation or test data and see which predicts the highest amount (response variable is revenue). Whichever model predicts the highest amount is my ending price segment. So if model A predicts the highest revenue for a customer, that customer is predicted to receive price A.

My model revenue is then the mean total revenue of all those customers who received the same price they were predicted to receive (or IOW all the customers who received the same price as the model that predicted the highest revenue)

I want to test different models to see which one predicts the highest model revenue. I can do it with sas code nodes manually, but I want to know if there is a way to do it automatically and select the highest models automatically.

I have learned that the start/end groups functionality wont work for my purposes since I cant ouput scores from each group with that functionality.

Any ideas? Thanks

Here is some help to show what I am doing:

Calculting uplift

Diagram of an example process (basically to show I am scoring the models separately. Each model is from a different subset of the data)

Re: Using start/end groups in Enterprise miner to obtain scores for each group

Xamius32 — Thu, 27 Oct 2016 15:55:28 GMT

I see in the score node that it scores each group seperately, IE if segment=1 then it scores one way, if segment=2 it scores another way. My goal is to have it score all observations X number of times (X=number of segments).

so, if in the score data, an observation is segment 1, it will score for segment 1,2 3, and 4. Thus I will have 4 different scores to compare.

Re: Using start/end groups in Enterprise miner to obtain scores for each group

Xamius32 — Thu, 27 Oct 2016 15:27:10 GMT

I appreciate the response. I had tried that originally, and when I put it before, the score node still only ends up with one score. I only see one prediction in the exported data under the score node as seen below.

Using start/end groups in Enterprise miner to obtain scores for each group

Xamius32 — Wed, 26 Oct 2016 22:50:21 GMT

I am trying to score data for each group selected in the start/end groups nodes in enterprise miner. The goal would be to gain a score for each group instead of just one score.

See example below. this example seems to end with one score, and I am not sure how to adjust to gain a score for each group for each observation in the validation data.

Re: Missing observations in scored data, no missing data

Xamius32 — Mon, 24 Oct 2016 21:35:18 GMT

So, I see that my socre node has different inputted data. One has the regression train data and one has the validation data, just not sure how that has happened.

Re: Missing observations in scored data, no missing data

Xamius32 — Mon, 24 Oct 2016 20:09:23 GMT

Well I am not sure that is the problem. Some of the scored data match the # of obs in the training data, and some match the # of obs in the validation data.I cant figure it out.

Re: Missing observations in scored data, no missing data

Xamius32 — Mon, 24 Oct 2016 19:07:28 GMT

Well I guess the scored data set will have more categories than the training data. Is that a problem?

Missing observations in scored data, no missing data

Xamius32 — Mon, 24 Oct 2016 18:16:07 GMT

I am building 4 different logistic models based on 4 different datasets, and then scoring 1 validation dataset with all 4 models to compare the scores.

But for some reason, the scored data is getting different # of observations, even though I know there are no missing values in training or validation data. Is there a reason this would occur?

Im trying to compare two variables and insert variable X into macro if conditions are met

Xamius32 — Thu, 25 Jun 2015 15:57:56 GMT

And create another dataset and add a row to the second dataset explaining the problem with X variable, so I any number of variables/explinations would need to be added to other dataset. So I am trying to figure out how to do this in an array (IE add 3 variables to the array and then add a row for each variable to say this variable has a problem)

My dataset looks like:

Variable	X	Y
A	1	2
B	1	.5

So the condition might be if X>Y then add a variable to array, and then I want to create dataset test and add a row saying variable X has X>Y

Is this possible and does it make sense

Trying to print everything from a dataset but format/style a few of the variables

Xamius32 — Wed, 13 May 2015 21:31:30 GMT

Basically, since some of the variables are so long they end up looking like scientific notation in excel, so I have to use the

var X /style(data)={tagattr='format:@'};

SO I want to print all the variables, but want to apply some styling options to just a few. How do I do that without writing var statements for all variables?

Re: Hi, I am new to IML, I am trying to add the reverse diagonals of each column

Xamius32 — Fri, 30 Jan 2015 21:34:08 GMT

Thinking about this more, it could be easier if I could figure out how to shift over rows by X amount. So for row 2, I would shift it over by 1, and the last observation would be deleted. Is that possible?

Hi, I am new to IML, I am trying to add the reverse diagonals of each column

Xamius32 — Fri, 30 Jan 2015 20:37:26 GMT

For example, Column 1 Row 1 would be a total. The Column 2 Row 1 plus Column 1 Row 2, then COL3 Row1+Col2 Row 2+ Col1 Row3, etc. For however many columns I have. It isnt a square matrix so once the columns exceed the rows it wouldnt match to the first column.

Header 1	Header 2	Header 3	Header 4	Header 5
	A	B	C	D
A	B	C	D
B	C	D

So for example above, I would want the totals of A, B, C, D etc.

I have the matrix, now just not sure how to add up what I need.

Any ideas?

Re: Trying to use first. and last. processing in 3rd variable of a by statement

Xamius32 — Tue, 09 Dec 2014 16:32:42 GMT

Thanks alot

Trying to use first. and last. processing in 3rd variable of a by statement

Xamius32 — Tue, 09 Dec 2014 16:10:35 GMT

For example, I am trying to see how often variable Z changes in the X group

X	Y	Z
A	201401	600
A	201402	600
A	201403	750

Variable X is in this example a particular loan, with Y being the monthend and Z being the score.

So lets say there are 10 identical Zs in a row and then the 11th observation is a different Z for the same X, I want to know that there were 10 in a row before changing.

I realize first.Z would be 1 for each Z because Y changes each observation (sorted by X Y Z).

I can get Z to add up correctly when I say if lag(Z)=Z then variable AA+1, but then I dont know how to output only the observation when AA changes back to one (when Z would have a new observation). Again, I cant use last. processing because Y changes each time, and we want the data in the order of Y.

Thanks for any help

Re: getting interactive decision tree in enterprise miner to use more than 20000 records

Xamius32 — Thu, 13 Mar 2014 18:57:41 GMT

I'm using 12.1. Where would those options be?