1
81%
master: 67%

Ran 06 Aug 2016 10:31AM UTC

Files 41

Run time 1s

Badge

Embed ▾

Committed 06 Aug 2016 10:28AM UTC coverage: 80.265%. Remained the same

Job # 1390.1

Build Type

push

travis-ci

Committed by

kortschak

Commit Message

community: validate expectation with louvain-igraph

Note that validation is subtle because of a bug in louvain igraph that
omits the quality of the first layer.

Do this:

assuming f and e are igraph graphs for the positive and negative layers

membership, quality = louvain.find_partition_multiplex([louvain.Layer(graph=f, method='Modularity', layer_weight=1.0), louvain.Layer(graph=e, method='Modularity', layer_weight=-1.0)], consider_comms=louvain.ALL_COMMS)
print membership
print louvain.quality(f, ig.VertexClustering(f, membership), method='Modularity')+quality

Also note that louvain-igraph underperforms the implementatio here and
many iterations are required to achieve the the score we get consistently.

Run Details

4238 of 5280 relevant lines covered (80.27%)

1.27 hits per line

gonum / graph / 1390 / 1
81%
master: 67%

README BADGES
x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 1390.1

gonum / graph / 1390 / 1 81% master: 67%

README BADGES x

Markdown

Textile

RDoc

HTML

Rst

Source Files on job 1390.1

gonum / graph / 1390 / 1
81%
master: 67%

README BADGES
x