Problem with generative_model (generative.py)

[pedrocklein]

Greetings,

I am running into an IndexError problem when using the evaluate_generative_model function.
To give you a better picture of what am I doing, I am using hyperopt to optimize the hyperparameters eta and gamma from the function evaluate_generative_model, and the parameter space for each is [-10, 10] in steps of 0.1, so for each trial, it selects two values from this space and send them to the function. Here is my code:

# Define the hyperparameters space
gamma = np.r_[-10:10:0.1]
eta = np.r_[-10:10:0.1]
hp_space = {
        'gamma': hp.choice('gamma', gamma),
        'eta': hp.choice('eta', eta)
}

# Setting up the number of evaluations and Trials object
n_evals = 2000
trials = Trials()

# Running the optimization
res_gnm = fmin(
            lambda hps: bct.evaluate_generative_model(seed_network,
                                                  target_network,
                                                  mean_dist,
                                                  [hps['eta']],
                                                  gamma=[hps['gamma']],
                                                  model_type='matching')[0],
            space=hp_space, algo=tpe.suggest, trials=trials, max_evals=n_evals,
            rstate=np.random.RandomState(42)
        )

Apparently, this problem disappears if I add this to the generative.py code (line 327)

...
for ii in range(mseed, m):
            C = np.append(0, np.cumsum(Ff[u,v]))
            r_n = np.sum(np.random.random()*C[-1] >= C) #storing the original index
            r = r_n - 1 #subtracting 1 from it
            uu = u[r]
            vv = v[r]
            A[uu,vv] = A[vv,uu] = 1
...

which is basically subtracting one from the index (I was assuming that it could be a 0-indexing problem, since the original code came from a matlab toolbox).

Even with this fix, though, the same error occurs when I use the evaluate_generative_model function passing eta and gamma as vectors.
gnm_eval = bct.evaluate_generative_model(seed_network, sc_dataset[0]['weights_bu'], mean_dist, eta, gamma=gamma)

Is there anything that I am doing wrong when using this function?

Thank you for your help.

P.S: I am attaching the dist, seed and target matrices as numpy files.
dist.txt
seed.txt
target.txt