Skip to content

fix: pass fname and correct nodetype kwarg when reading weighted graphs - #8

Open
JOhnsonKC201 wants to merge 1 commit into
mims-harvard:masterfrom
JOhnsonKC201:fix/weighted-read-edgelist
Open

fix: pass fname and correct nodetype kwarg when reading weighted graphs#8
JOhnsonKC201 wants to merge 1 commit into
mims-harvard:masterfrom
JOhnsonKC201:fix/weighted-read-edgelist

Conversation

@JOhnsonKC201

Copy link
Copy Markdown

Fixes #5.

read_net's weighted branch calls:

G = nx.read_edgelist(inodetype=int, data=(('weight', float),),
                     create_using=nx.DiGraph())

This never passes the edgelist path and hands read_edgelist an inodetype keyword it doesn't accept, so any run with --weighted dies immediately with:

TypeError: read_edgelist() got an unexpected keyword argument 'inodetype'

before any embedding work happens. The fix passes fname as the path and uses nodetype=int, matching the unweighted branch right below it. Unweighted runs are unaffected.

@marinkaz offered to merge a fix for this in #5; #6 made the same change but was auto-closed when the contributor's fork was removed, so this picks it back up.

read_net's weighted branch called nx.read_edgelist(inodetype=int, ...),
which omits the required edgelist path and passes an argument
read_edgelist does not accept. Any run with --weighted aborts immediately
with 'TypeError: read_edgelist() got an unexpected keyword argument
inodetype' before any embedding work happens.

Pass fname as the path and use nodetype=int, matching the unweighted
branch just below. Closes mims-harvard#5.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Weighted graph mistakes

1 participant