community
cancel
Showing results for 
Search instead for 
Did you mean: 

Alteryx Designer Ideas

Share your Designer product ideas - we're listening!
New Data Science Blog

Check out the latest post: All Models Are Wrong

READ MORE
Announcement | We'll be doing maintenance between pm 4- 6pm MT, which may impact your experience. Thanks for your patience as we work on improving the community!

High speed regression

I am trying to run batch regressions on a pretty sizable set of data.  About ~1M distinct groups of data, each wtih 30-500 x,y pairs.

 

A batch macro with a linear regression works ok - but it is really slow.  Started at about 2-3s per regression.  After stripping out bunch or reporting from the macro, I am down to ~2s.  This is still feels quite slow compared to something purpose built.

 

Has anyone experimented with higher speed versions that just dump out m,b, & r2?

2 Comments
Asteroid

I had a similar issue.. the most amount of time is consumed in opeing R instance for the regression for every iteration. I worked around by- instead of using regression tool, use R tool and write your regression with Group by argument in there.

Alteryx Partner

Actually what we need seems to be a very similar case with TS model factory...

But this time "Regression model factory"...