search

naobservatory / p2ra

4 stars 1 forks source link

Set pseudocounts back to 0.1 infections per week (previously 0.001). Reran fit.py. #238

Closed simonleandergrimm closed 9 months ago

simonleandergrimm commented 9 months ago

I noticed that our pseudocount value was too low (0.001 infections per week), even though it should be 0.1 infections per week. I failed to spot this in #229, where, as part of a sensitivity analysis, 0.1 was changed to 0.001 and subsequently committed to main. I now fixed this.

I reran run.py, which created a new fits_summary.tsv. @jeffkaufman Once you approve these changes I can replace the violin figures in the P2RA report

jeffkaufman commented 9 months ago

Thanks for catching and fixing my mistake!