BookmarkSubscribeRSS Feed
Ullsokk
Pyrite | Level 9

I have a gradient boosting model that I want to to do some LIME on. But we don't have Viya yet. I have tried using python, but it seems I have to run the model in python as well, so the model I get LIME plots from is not identical to my SAS model. 

 

Is there a way to run LIME in sas without Viya?

2 REPLIES 2
RalphAbbey
SAS Employee

Unfortunately to my knowledge the only way would be to write a SAS macro that performs the steps of LIME yourself. This might be a little short on details, but cover these steps cover tabular LIME:

 

1) Run an analysis to obtain the variance for each input variable

2) Choose an observation which you wish to run LIME on

3) Create N samples, where the observations are sampled from a Gaussian normal distribution with mean equal to the variable value for observation from 2) and the variance equal to the variance from 1)

4) Calculate weights for each observation using the RBF kernel between the observation in 2) and the generated observations in 3)

5) Score this new data set using the model

6) Run a weighted regression on the scored data set using the output of the model as the dependent variable and the inputs of the model as the independent variables

 

Let me know if this helps.

Ullsokk
Pyrite | Level 9

Thank you! It helps knowing there is no easy sollution. Looks like I will be running the models in python until we are on Viya. Implementing LIME by hand does sound a bit excessive, can't risk spending that much work on something I am not even sure I can pull off.

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
How to choose a machine learning algorithm

Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 2 replies
  • 774 views
  • 0 likes
  • 2 in conversation