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
Why would the PC implemented in pc.py in the gCastle package strictly use rule 4 even without background knowledge?
# rule4
for i, j in permutations(ij, 2):
for kj in sep_set.keys(): # k and j are nonadjacent.
if j not in kj:
continue
else:
kj = list(kj)
kj.remove(j)
k = kj[0]
ls = [x for x in columns if x not in [i, j, k]]
for l in ls:
if cpdag[k, l] == 1 \
and cpdag[l, k] == 0 \
and cpdag[i, k] == 1 \
and cpdag[k, i] == 1 \
and cpdag[l, j] == 1 \
and cpdag[j, l] == 0:
cpdag[j, i] = 0
One more question, don't we need to have cpdag[l, i] == 1 and cpdag[i, l] ==1 in the if condition?
The text was updated successfully, but these errors were encountered:
Hi,
pc.py
in thegCastle
package strictly use rule 4 even without background knowledge?cpdag[l, i] == 1 and cpdag[i, l] ==1
in the if condition?The text was updated successfully, but these errors were encountered: