CBIIT / R-cometsAnalytics

R package development for COMETS Analytics
12 stars 10 forks source link

COMETS 1.5. Run time is long when no covariates used for adjustment #53

Closed steven-moore closed 1 year ago

steven-moore commented 6 years ago

The "Run model" is checking the data file again on the website end but it doesn't need to--the data is already saved in the created list and everything is there.

Resolving this will help speed the analyses.

steven-moore commented 6 years ago

It turns out that this is not an issue of loading the dataset, but rather the adjustments (which I had tended to run in a particular order). The analysis time for a model of age*all metabolites, with no adjustments is about 32 second.

image

However, if I add bmi_grp to the model, analysis time drops to <5 seconds. This is true even when I logout and login and run models fresh. This issue seems minor to me, but is perhaps worth brief investigation to determine why the algorithm gets bogged down when there are no covariates in the model.

image

steven-moore commented 6 years ago

Ella will investigate why the algorithm takes longer when no adjustments used. To my thinking, the hang up is most likely in the model checks, e.g. CARET, where the algorithm is looking for adjustments but not finding any.

steven-moore commented 5 years ago

Ella, any progress on this particular issue?