sphinxteam / sir_inference Goto Github PK

There's an error in model = ProximityModel(N=N, scale=1.2, mu=0.01, lamb=0.02 ,ground_truth=ground_truth) as the argument ground_truth is missing from the ProximityModel class.

Error Message:

----> model = ProximityModel(N=N, scale=1.2, mu=0.01, lamb=0.02 ,ground_truth=ground_truth)
TypeError: __init__() got an unexpected keyword argument 'ground_truth'

Notebook Scatterplot_mean_field_dmp uses the model instance two times (for two scatterplot sets) to generate initial_probas. However, initial state of the model is changed to a random state in the first run because generate_scatterplot calls compute_averages which sets the initial states to random.

This might make the two scatterplot sets incomparable (I am not sure about this, haven't fully read through the code, just noticed this inconsistency).