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
another question about lmerge:213
sum_L = sum(p_L)
sum_R = sum(p_R) not used
p_L = [x/sum_L for x in p_L]
p_R = [x/sum_L for x in p_R] sum_L used twice
on this line sum_R is not used and sum_L is used both on R and L ,is this what we wanted?
we have 1000 samples and when lmerge after lsort, lmerge:213 report divide zero error.
some debug info:
combine_pdfs(BP, c ,use_product, weighting_scheme)
a_L[0] p_L[0] on lmerge:128
miss=1 on lmerge:181
I doubt it may be caused by the "unweighted" scheme? and all the BP has no internals?
The text was updated successfully, but these errors were encountered: