preprocessing.py

import numpy as np
import scipy.sparse as sp


def sparse_to_tuple(sparse_mx):
    if not sp.isspmatrix_coo(sparse_mx):
        sparse_mx = sparse_mx.tocoo()
    coords = np.vstack((sparse_mx.row, sparse_mx.col)).transpose()
    values = sparse_mx.data
    shape = sparse_mx.shape
    return coords, values, shape


def preprocess_graph(adj):
    adj = sp.coo_matrix(adj)
    adj_ = adj + sp.eye(adj.shape[0])
    rowsum = np.array(adj_.sum(1))
    degree_mat_inv_sqrt = sp.diags(np.power(rowsum, -0.5).flatten())
    adj_normalized = adj_.dot(degree_mat_inv_sqrt).transpose().dot(degree_mat_inv_sqrt).tocoo()
    return sparse_to_tuple(adj_normalized)


def construct_feed_dict(B_1,adj_normalized, adj, features, placeholders,B):
    # construct feed dictionary
    feed_dict = dict()
    feed_dict.update({placeholders['features']: features})
    feed_dict.update({B_1: B})
    feed_dict.update({placeholders['adj']: adj_normalized})
    feed_dict.update({placeholders['adj_orig']: adj})
    return feed_dict