search

weizhou0 / qtl

GNU General Public License v3.0
21 stars 2 forks source link

Reduce step1 model file size #8

Closed mkanai closed 3 months ago

mkanai commented 4 months ago

I added a new option --isShrinkModelOutput (default: TRUE) to reduce step1 model file size. Unnecessary objects in modglmm (which becomes NULL in ReadModel to save memory) are removed before save.

This option reduces .rda file size by 50-60% when eCovars are not included

mkanai commented 3 months ago

As discussed, Sigma_iXXSigma_iX is now removed from the step1. Currently, it only takes into effect when --isShrinkModelOutput=TRUE just in case it might be required for future gxe. This now reduces .rda by 80-90%