Add the possibility to generate networks with random number of nodes

[danielemule]

It is useful to give the user the possibility to create a sample of networks with random number of nodes, in particular: N_1,N_2,...,N_{sample_size} iid Poisson(num_vertices).

[danielemule]

I have added this possibility to nvd(), via the rand_num_vertices argument.

You can find it in my graphmatch branch of nevada (link).

[danielemule]

I have added the possibility to generate networks with random number of nodes following an Uniform distribution.

You can find it in my graphmatch branch of nevada (link).

[astamm]

Please do open separate pull request for each of the issues instead of the links. Thanks!

[astamm]

I think in this case you need to use the play_*() functions (either the ones in tidygraph or the ones in nevada) which are specialised in generating one graph.

You can first generate a vector of integers that give you the number of vertices for each graph. And then a graph with the appropriate number of vertices for each entry of that vector.

The reason you need to do this by hand is that currently there is no unified way (naming convention) to specify the number of vertices across all play_*() functions. Some of them even infer it using the length of other provided parameters (see play_degree() for instance). I am not sure this is something @thomasp85 has in the roadmap for tidygraph.