Typo in docs: changes results?
In the edge covariate example, when you pad the state for one of the examples I think you're using the wrong state variable? https://git.skewed.de/count0/graph-tool/-/blob/master/doc/demos/inference/_edge_weights.rst#L227
state_ln = state.copy(bs=state.get_bs() + [np.zeros(1)] * 4, sampling=True)
Should be ?
state_ln = state_ln.copy(bs=state_ln.get_bs() + [np.zeros(1)] * 4, sampling=True)
With this change I'm getting a significantly differently posterior odds ratio:
ln Λ: -30.50104261051456