@signehaahr (please reply to this issue when you have read your feedback)
The warmup
Your proof is only true for q being a natural number, this is a very important detail! Other than that your solutions looks o.k. You are missing half of the solutions because you only find the ones where a<b. There are a lot of mirrored solutions b,a for each of your solutions a,b. You dont need to compute them, you can simply plot another scatter plot with the coordinates flipped. The patterns might also become more clear if you reduce the marker size a bit in the figure.
The exercises
Looks very good ! In your modularity function you dont need to rewrite all of the code you have just written for the delta function. You can call delta(ci,cj) inside the modularity function.
Really good job with the louvain method. It is clear that you have taken inspiration from the writeup but modified the code to something you fully know how works.
signehaahr
commented
4 years ago
@signehaahr (please reply to this issue when you have read your feedback)
The warmup Your proof is only true for q being a natural number, this is a very important detail! Other than that your solutions looks o.k. You are missing half of the solutions because you only find the ones where a<b. There are a lot of mirrored solutions b,a for each of your solutions a,b. You dont need to compute them, you can simply plot another scatter plot with the coordinates flipped. The patterns might also become more clear if you reduce the marker size a bit in the figure.
The exercises Looks very good ! In your modularity function you dont need to rewrite all of the code you have just written for the
deltafunction. You can calldelta(ci,cj)inside the modularity function.Really good job with the louvain method. It is clear that you have taken inspiration from the writeup but modified the code to something you fully know how works.
Score: 1