:quality(75))
Key takeaways
- Stanford math and data science intern Anqi Fu was characterizing GLMNet in R on sparse data using a HUD dataset predicting two-bedroom median rent by neighborhood.
- The post said the analysis initially ran quickly, until checking the data matrix and factor call revealed that most of the data was missing.
- The post said results of the intern's planned follow-up work, 'Characterizing GLMNet on Sparse Matrices,' would come in a future post.
All in the day:
Anqi Fu, our wickedly smart Math & Data Science hacker-intern from Stanford this summer, was characterizing GLMNet in R on sparse data and comparing with other tools. We were using a data sets predicting Two Bedroom median rent based on neighborhoods from huduser.org.
DATA: http://www.huduser.org/portal/datasets/fmr/CensusRentData/index.html
She found the analysis brisk and surprisingly fast.. Until we got around to checking the data matrix and the factor call. Most of the data was missing! So she exclaimed:
[Credits to Addletters.org & Matt Groenig for the Simpsons]
Results of her work “Characterizing GLMNet on Sparse Matrices”, will have to wait for a future post!