d5398270501cd5d6e9880db5e15ebf407ddec5ee,examples/plot_expectation_maximisation_manifolds.py,,weighted_gmm_pdf,#Any#Any#Any#Any#Any#,160

Before Change

def weighted_gmm_pdf(w, z, mu, sigma, distance):
    // print(z.size())
    // print(z.size(0), len(mu), z.size(1))
    z_u = z.unsqueeze(1).expand(z.size(0), len(mu), z.size(1))
    // print(z_u.size())
    // print(mu.size())
    mu_u = mu.unsqueeze(0).expand_as(z_u)

    distance_to_mean = distance(z_u, mu_u)
    sigma_u = sigma.unsqueeze(0).expand_as(distance_to_mean)
    distribution_normal = torch.exp(-((distance_to_mean)**2)/(2 * sigma_u**2))

After Change

    mesh_data_units = gs.expand_dims(mesh_data, 1)

    mesh_data_units = gs.repeat(mesh_data_units, len(means), axis = 1)

    //means_units = gs.expand_dims(means,0)

    means_units = gs.expand_dims(means,0)


    means_units = gs.repeat(means_units,mesh_data_units.shape[0],axis = 0)
    mesh_data_units = torch.from_numpy(mesh_data_units)
    means_units = torch.from_numpy(means_units)


    //means_units = means_units.expand_as(mesh_data_units)

Instances