search

stoufferlab / pymfinder

A Python package with which to find network motifs in complex networks and to analyze a growing list of network-motif related stuff.
MIT License
16 stars 7 forks source link

Added a new test for bipartite networks #8

Closed bernibra closed 8 years ago

bernibra commented 8 years ago

I've generated networks containing a single motif of each type for all five possible sizes (i.e. five networks) and I've added a new test checking that all the different species in those networks play different roles.

stouffer commented 8 years ago

Sounds like a fantastic way to check things. One question though, shouldn't the roles of A1 and A2, and B1 and B2, be the same in bipartite-2-test.net?

I may just be confused by the way the test works as there are lots of oddball list comprehensions that I'm unfamiliar with.

bernibra commented 8 years ago

They are, but I think that I got an error when I had a single interaction. It is something I wanted to doublecheck tomorrow. El dia 02/03/2016 19:09, "Daniel B. Stouffer" notifications@github.com va escriure:

Sounds like a fantastic way to check things. One question though, shouldn't the roles of A1 and A2, and B1 and B2, be the same in bipartite-2-test.net?

— Reply to this email directly or view it on GitHub https://github.com/stoufferlab/pymfinder/pull/8#issuecomment-191082918.

bernibra commented 8 years ago

It is up to 6. The networks are size 2, 3, 4, 5 and 6. El dia 02/03/2016 19:54, "Daniel B. Stouffer" notifications@github.com va escriure:

Plus, why only up to 5 since the code should work up to 6? I would totally get it if it was a lack of time/patience to add that one, but it does seem like an important to-do item.

— Reply to this email directly or view it on GitHub https://github.com/stoufferlab/pymfinder/pull/8#issuecomment-191097034.

stouffer commented 8 years ago

Yes, I noticed in the end that it did actually go to six (and deleted that comment). I got thrown off by the fact that the later loop was from 0 to 5 and not 2 to 7 and hadn't thought about it long enough.

As far as an error goes, you probably had an empty line. I just tried with a single interaction and it passes the test fine.

bernibra commented 8 years ago

You are right, I did it at the very last moment today and I didn't think too much about it! I'll change it tomorrow! El dia 02/03/2016 22:18, "Daniel B. Stouffer" notifications@github.com va escriure:

Yes, I noticed in the end that it did actually go to six (and deleted that comment). I got thrown off by the fact that the later loop was from 0 to 5 and not 2 to 7 and hadn't thought about it long enough.

As far as an error goes, you probably had an empty line. I just tried with a single interaction and it passes the test fine.

— Reply to this email directly or view it on GitHub https://github.com/stoufferlab/pymfinder/pull/8#issuecomment-191146537.