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
{{ message }}
This repository has been archived by the owner on Apr 22, 2020. It is now read-only.
I've been searching the docs, but haven't found anything on this. If I oversaw something, please link me there :)
I'm using version 3.5.4.1.
I tried writing my own implementation of shortestPath.deltaStepping and noticed in your tests that you are using three kernel types, i.e. light, heavy, kernel.
I don't quite understand their functions. I understand that you load a temporary projection of the database into memory and those graph types define how that projection looks like. What differences do they make? Where are their characteristics defined?
I have found the definition of the graph type hugehere.
From experiments with my custom procedure, I noticed that light, heavy do not keep parallel relationships between a pair of nodes by default, but kernel does.
Also, using huge with duplicateRelationships: 'none', contrary to the docs, seem to deduplicate relationships, based on my experiments.
My experiments have been kind of pseudo and hence I'm not sharing them here. I am just looking for the exact definitions of the mentioned graph types. kernel does seem to do what I want, that is keeping parallel relationships between a pair of nodes, however I cannot find its documentation and thus I'm afraid of using it in production.
Thank you in advance!
The text was updated successfully, but these errors were encountered:
Sign up for freeto subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Hi,
I've been searching the docs, but haven't found anything on this. If I oversaw something, please link me there :)
I'm using
version 3.5.4.1
.I tried writing my own implementation of
shortestPath.deltaStepping
and noticed in your tests that you are using three kernel types, i.e.light
,heavy
,kernel
.I don't quite understand their functions. I understand that you load a temporary projection of the database into memory and those graph types define how that projection looks like. What differences do they make? Where are their characteristics defined?
I have found the definition of the graph type
huge
here.From experiments with my custom procedure, I noticed that
light
,heavy
do not keep parallel relationships between a pair of nodes by default, butkernel
does.Also, using
huge
withduplicateRelationships: 'none'
, contrary to the docs, seem to deduplicate relationships, based on my experiments.My experiments have been kind of pseudo and hence I'm not sharing them here. I am just looking for the exact definitions of the mentioned graph types.
kernel
does seem to do what I want, that is keeping parallel relationships between a pair of nodes, however I cannot find its documentation and thus I'm afraid of using it in production.Thank you in advance!
The text was updated successfully, but these errors were encountered: