You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm trying to accelerate minimap2 on GPU. I noticed some inconsistency between my implementation of ksw2 with bandwidth w!= inf which I followed the algorithm in the minimap2 paper, and the implementation living on master. I also notice intel's mm2-fast implementation cause inconsistency as well. I'm trying to dig into the root cause of this and suspect the initialization might be it.
I want to confirm my understanding of the initialization process: We need to initialize both green and red cells while yellow are the cells we calculate within the band. Green cells are initialized to
in the minimap2 paper. Red cells are initialized to u = v = -q-e, x=y=-q-e and x2=y2=-q2-e2 according to the code?
The text was updated successfully, but these errors were encountered:
I'm trying to accelerate minimap2 on GPU. I noticed some inconsistency between my implementation of ksw2 with bandwidth w!= inf which I followed the algorithm in the minimap2 paper, and the implementation living on master. I also notice intel's mm2-fast implementation cause inconsistency as well. I'm trying to dig into the root cause of this and suspect the initialization might be it.


I want to confirm my understanding of the initialization process: We need to initialize both green and red cells while yellow are the cells we calculate within the band. Green cells are initialized to
in the minimap2 paper. Red cells are initialized to
u = v = -q-e
,x=y=-q-e
andx2=y2=-q2-e2
according to the code?The text was updated successfully, but these errors were encountered: