utils.py

import numpy as np
import pickle as pkl
import networkx as nx
import scipy.sparse as sp
from scipy.sparse.linalg.eigen.arpack import eigsh
import sys


def parse_index_file(filename):
    """Parse index file."""
    index = []
    for line in open(filename):
        index.append(int(line.strip()))
    return index


def sample_mask(idx, l):
    """Create mask."""
    mask = np.zeros(l)
    mask[idx] = 1
    return np.array(mask, dtype=np.bool)


def load_data(dataset_str):
    """
    Loads input data from gcn/data directory

    ind.dataset_str.x => the feature vectors of the training instances as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.tx => the feature vectors of the test instances as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.allx => the feature vectors of both labeled and unlabeled training instances
        (a superset of ind.dataset_str.x) as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.y => the one-hot labels of the labeled training instances as numpy.ndarray object;
    ind.dataset_str.ty => the one-hot labels of the test instances as numpy.ndarray object;
    ind.dataset_str.ally => the labels for instances in ind.dataset_str.allx as numpy.ndarray object;
    ind.dataset_str.graph => a dict in the format {index: [index_of_neighbor_nodes]} as collections.defaultdict
        object;
    ind.dataset_str.test.index => the indices of test instances in graph, for the inductive setting as list object.

    All objects above must be saved using python pickle module.

    :param dataset_str: Dataset name
    :return: All data input files loaded (as well the training/test data).
    """
    names = ['x', 'y', 'tx', 'ty', 'allx', 'ally', 'graph']
    objects = []
    for i in range(len(names)):
        with open("data/ind.{}.{}".format(dataset_str, names[i]), 'rb') as f:
            if sys.version_info > (3, 0):
                objects.append(pkl.load(f, encoding='latin1'))
            else:
                objects.append(pkl.load(f))

    x, y, tx, ty, allx, ally, graph = tuple(objects)
    test_idx_reorder = parse_index_file("data/ind.{}.test.index".format(dataset_str))
    test_idx_range = np.sort(test_idx_reorder)

    if dataset_str == 'citeseer':
        # Fix citeseer dataset (there are some isolated nodes in the graph)
        # Find isolated nodes, add them as zero-vecs into the right position
        test_idx_range_full = range(min(test_idx_reorder), max(test_idx_reorder)+1)
        tx_extended = sp.lil_matrix((len(test_idx_range_full), x.shape[1]))
        tx_extended[test_idx_range-min(test_idx_range), :] = tx
        tx = tx_extended
        ty_extended = np.zeros((len(test_idx_range_full), y.shape[1]))
        ty_extended[test_idx_range-min(test_idx_range), :] = ty
        ty = ty_extended

    ##########################################################
    # reorder because tx and ty are not ordered w.r.t. node ID
    ##########################################################
    features = sp.vstack((allx, tx)).tolil()
    features[test_idx_reorder, :] = features[test_idx_range, :]
    adj = nx.adjacency_matrix(nx.from_dict_of_lists(graph))

    labels = np.vstack((ally, ty))
    labels[test_idx_reorder, :] = labels[test_idx_range, :]

    idx_test = test_idx_range.tolist()
    idx_train = range(len(y))
    idx_val = range(len(y), len(y)+500)

    train_mask = sample_mask(idx_train, labels.shape[0])
    val_mask = sample_mask(idx_val, labels.shape[0])
    test_mask = sample_mask(idx_test, labels.shape[0])

    y_train = np.zeros(labels.shape)
    y_val = np.zeros(labels.shape)
    y_test = np.zeros(labels.shape)
    y_train[train_mask, :] = labels[train_mask, :]
    y_val[val_mask, :] = labels[val_mask, :]
    y_test[test_mask, :] = labels[test_mask, :]

    return adj, features, y_train, y_val, y_test, train_mask, val_mask, test_mask


def load_data_gen(dataset_str):
    """
    Loads input data from gcn/data directory

    ind.dataset_str.x => the feature vectors of the training instances as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.tx => the feature vectors of the test instances as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.allx => the feature vectors of both labeled and unlabeled training instances
        (a superset of ind.dataset_str.x) as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.y => the one-hot labels of the labeled training instances as numpy.ndarray object;
    ind.dataset_str.ty => the one-hot labels of the test instances as numpy.ndarray object;
    ind.dataset_str.ally => the labels for instances in ind.dataset_str.allx as numpy.ndarray object;
    ind.dataset_str.graph => a dict in the format {index: [index_of_neighbor_nodes]} as collections.defaultdict
        object;
    ind.dataset_str.test.index => the indices of test instances in graph, for the inductive setting as list object.

    All objects above must be saved using python pickle module.

    :param dataset_str: Dataset name
    :return: All data input files loaded (as well the training/test data).
    """
    names = ['x', 'y', 'tx', 'ty', 'allx', 'ally', 'graph']
    objects = []
    for i in range(len(names)):
        with open("data/ind.{}.{}".format(dataset_str, names[i]), 'rb') as f:
            if sys.version_info > (3, 0):
                objects.append(pkl.load(f, encoding='latin1'))
            else:
                objects.append(pkl.load(f))

    x, y, tx, ty, allx, ally, graph = tuple(objects)
    test_idx_reorder = parse_index_file("data/ind.{}.test.index".format(dataset_str))
    test_idx_range = np.sort(test_idx_reorder)

    if dataset_str == 'citeseer':
        # Fix citeseer dataset (there are some isolated nodes in the graph)
        # Find isolated nodes, add them as zero-vecs into the right position
        test_idx_range_full = range(min(test_idx_reorder), max(test_idx_reorder)+1)
        tx_extended = sp.lil_matrix((len(test_idx_range_full), x.shape[1]))
        tx_extended[test_idx_range-min(test_idx_range), :] = tx
        tx = tx_extended
        ty_extended = np.zeros((len(test_idx_range_full), y.shape[1]))
        ty_extended[test_idx_range-min(test_idx_range), :] = ty
        ty = ty_extended

    ##########################################################
    # reorder because tx and ty are not ordered w.r.t. node ID
    ##########################################################
    features = sp.vstack((allx, tx)).tolil()
    features[test_idx_reorder, :] = features[test_idx_range, :]
    adj = nx.adjacency_matrix(nx.from_dict_of_lists(graph))

    labels = np.vstack((ally, ty))
    labels[test_idx_reorder, :] = labels[test_idx_range, :]

    idx_test = test_idx_range.tolist()
    idx_train = range(len(y))
    idx_val = range(len(y), len(y)+500)

    train_mask = sample_mask(idx_train, labels.shape[0])
    val_mask = sample_mask(idx_val, labels.shape[0])
    test_mask = sample_mask(idx_test, labels.shape[0])

    y_train = np.zeros(labels.shape)
    y_val = np.zeros(labels.shape)
    y_test = np.zeros(labels.shape)
    y_train[train_mask, :] = labels[train_mask, :]
    y_val[val_mask, :] = labels[val_mask, :]
    y_test[test_mask, :] = labels[test_mask, :]

    return adj, features, y_train, y_val, y_test, train_mask, val_mask, test_mask, labels


def load_data_split(dataset_str, train_ratio=0, repeat=0):
    """
    Loads input data from gcn/data directory

    ind.dataset_str.x => the feature vectors of the training instances as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.tx => the feature vectors of the test instances as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.allx => the feature vectors of both labeled and unlabeled training instances
        (a superset of ind.dataset_str.x) as scipy.sparse.csr.csr_matrix object;
    ind.dataset_str.y => the one-hot labels of the labeled training instances as numpy.ndarray object;
    ind.dataset_str.ty => the one-hot labels of the test instances as numpy.ndarray object;
    ind.dataset_str.ally => the labels for instances in ind.dataset_str.allx as numpy.ndarray object;
    ind.dataset_str.graph => a dict in the format {index: [index_of_neighbor_nodes]} as collections.defaultdict
        object;
    ind.dataset_str.test.index => the indices of test instances in graph, for the inductive setting as list object.

    All objects above must be saved using python pickle module.

    :param dataset_str: Dataset name
    :return: All data input files loaded (as well the training/test data).
    """
    names = ['x', 'y', 'tx', 'ty', 'allx', 'ally', 'graph']
    objects = []
    for i in range(len(names)):
        with open("data/ind.{}.{}".format(dataset_str, names[i]), 'rb') as f:
            if sys.version_info > (3, 0):
                objects.append(pkl.load(f, encoding='latin1'))
            else:
                objects.append(pkl.load(f))

    x, y, tx, ty, allx, ally, graph = tuple(objects)
    test_idx_reorder = parse_index_file("data/ind.{}.test.index".format(dataset_str))
    test_idx_range = np.sort(test_idx_reorder)

    if dataset_str == 'citeseer':
        # Fix citeseer dataset (there are some isolated nodes in the graph)
        # Find isolated nodes, add them as zero-vecs into the right position
        test_idx_range_full = range(min(test_idx_reorder), max(test_idx_reorder)+1)
        tx_extended = sp.lil_matrix((len(test_idx_range_full), x.shape[1]))
        tx_extended[test_idx_range-min(test_idx_range), :] = tx
        tx = tx_extended
        ty_extended = np.zeros((len(test_idx_range_full), y.shape[1]))
        ty_extended[test_idx_range-min(test_idx_range), :] = ty
        ty = ty_extended

    ##########################################################
    # reorder because tx and ty are not ordered w.r.t. node ID
    ##########################################################
    features = sp.vstack((allx, tx)).tolil()
    features[test_idx_reorder, :] = features[test_idx_range, :]
    adj = nx.adjacency_matrix(nx.from_dict_of_lists(graph))

    labels = np.vstack((ally, ty))
    labels[test_idx_reorder, :] = labels[test_idx_range, :]

    # if train_ratio == 0:
    #     idx_test = test_idx_range.tolist()
    #     idx_train = range(len(y))
    #     idx_val = range(len(y), len(y)+500)
    # else:
    #     train_count = train_ratio * labels.shape[0] // 100
    #     idx_train = range(train_count)
    #     idx_test = []
    #     idx_val = []
    #     for i in range(train_count, labels.shape[0]):
    #         if i % 2 == 0:
    #             idx_test.append(i)
    #         else:
    #             idx_val.append(i)
    # print('#train instances:', len(idx_train))
    # print('#val instances:', len(idx_val))
    # print('#test instances:', len(idx_test))
    #
    # train_mask = sample_mask(idx_train, labels.shape[0])
    # val_mask = sample_mask(idx_val, labels.shape[0])
    # test_mask = sample_mask(idx_test, labels.shape[0])

    # load flags
    with open("data/ind.{}.{}.{}.flag".format(dataset_str, str(train_ratio),
                                              str(repeat)), 'rb') as f:
        if sys.version_info > (3, 0):
            sample_flags = pkl.load(f, encoding='latin1')
        else:
            sample_flags = pkl.load(f)

    train_mask = sample_flags == 1
    val_mask = sample_flags == 2
    test_mask = sample_flags == 3

    y_train = np.zeros(labels.shape)
    y_val = np.zeros(labels.shape)
    y_test = np.zeros(labels.shape)
    y_train[train_mask, :] = labels[train_mask, :]
    y_val[val_mask, :] = labels[val_mask, :]
    y_test[test_mask, :] = labels[test_mask, :]

    # print(np.sum(labels, axis=0))
    # print(np.sum(y_train, axis=0))
    # print(np.sum(y_val, axis=0))
    # print(np.sum(y_test, axis=0))

    return adj, features, y_train, y_val, y_test, train_mask, val_mask, test_mask


def load_data_reddit(dataset_str, train_ratio=0, repeat=0):
    """
    Loads input data from gcn/data directory

    ind.dataset_str.feats => the feature vectors of all nodes as 2d numpy array;
    ind.dataset_str.labs => the one hot label vectors of all nodes as 2d numpy array;
    ind.dataset_str.tr_ids => indices of training nodes as lis;
    ind.dataset_str.va_ids => indices of validation nodes as lis;
    ind.dataset_str.te_ids => indices of testing nodes as lis;
    ind.dataset_str.graph => a dict in the format {index: [index_of_neighbor_nodes]}
        as collections.defaultdict object;

    All objects above must be saved using python pickle module.

    :param dataset_str: Dataset name
    :return: All data input files loaded (as well the training/test data).
    """
    names = ['feats', 'labs', 'tr_ids', 'va_ids', 'te_ids', 'graph']
    objects = []
    for i in range(len(names)):
        with open("data/ind.{}.{}".format(dataset_str, names[i]), 'rb') as f:
            if sys.version_info > (3, 0):
                objects.append(pkl.load(f, encoding='latin1'))
            else:
                objects.append(pkl.load(f))

    features, labels, idx_train, idx_val, idx_test, graph = tuple(objects)

    adj = nx.adjacency_matrix(nx.from_dict_of_lists(graph))

    # if not train_ratio == 0:
    #     train_count = train_ratio * labels.shape[0] // 100
    #     idx_train = range(train_count)
    #     idx_test = []
    #     idx_val = []
    #     for i in range(train_count, labels.shape[0]):
    #         if i % 2 == 0:
    #             idx_test.append(i)
    #         else:
    #             idx_val.append(i)
    # print('#train instances:', len(idx_train))
    # print('#val instances:', len(idx_val))
    # print('#test instances:', len(idx_test))
    #
    # train_mask = sample_mask(idx_train, labels.shape[0])
    # val_mask = sample_mask(idx_val, labels.shape[0])
    # test_mask = sample_mask(idx_test, labels.shape[0])

    # load flags
    with open("data/ind.{}.{}.{}.flag".format(dataset_str, str(train_ratio),
                                              str(repeat)), 'rb') as f:
        if sys.version_info > (3, 0):
            sample_flags = pkl.load(f, encoding='latin1')
        else:
            sample_flags = pkl.load(f)

    train_mask = sample_flags == 1
    val_mask = sample_flags == 2
    test_mask = sample_flags == 3

    y_train = np.zeros(labels.shape)
    y_val = np.zeros(labels.shape)
    y_test = np.zeros(labels.shape)
    y_train[train_mask, :] = labels[train_mask, :]
    y_val[val_mask, :] = labels[val_mask, :]
    y_test[test_mask, :] = labels[test_mask, :]

    # print(np.sum(labels, axis=0))
    # print(np.sum(y_train, axis=0))
    # print(np.sum(y_val, axis=0))
    # print(np.sum(y_test, axis=0))

    return adj, features, y_train, y_val, y_test, train_mask, val_mask, test_mask


def sparse_to_tuple(sparse_mx):
    """Convert sparse matrix to tuple representation."""
    def to_tuple(mx):
        if not sp.isspmatrix_coo(mx):
            mx = mx.tocoo()
        coords = np.vstack((mx.row, mx.col)).transpose()
        values = mx.data
        shape = mx.shape
        return coords, values, shape

    if isinstance(sparse_mx, list):
        for i in range(len(sparse_mx)):
            sparse_mx[i] = to_tuple(sparse_mx[i])
    else:
        sparse_mx = to_tuple(sparse_mx)

    return sparse_mx


def synthetic_features(sample_num, labels, syn_classes=0):
    # sample from standard normal distribution
    print(np.sum(labels, axis=0))
    syn_feas = np.random.normal(0, 1, [sample_num, labels.shape[1] * 2])
    for i in range(syn_classes):
        for j in range(sample_num):
            if labels[j][i] > 0.5:
                # print(i, j)
                if syn_feas[j][i * 2] * syn_feas[j][i * 2 + 1] <= 0:
                    if np.random.randint(2) == 0:
                        syn_feas[j][i * 2] *= -1
                    else:
                        syn_feas[j][i * 2 + 1] *= -1
            else:
                if syn_feas[j][i * 2] * syn_feas[j][i * 2 + 1] > 0:
                    if np.random.randint(2) == 0:
                        syn_feas[j][i * 2] *= -1
                    else:
                        syn_feas[j][i * 2 + 1] *= -1
    return syn_feas


def preprocess_features(features):
    """Row-normalize feature matrix and convert to tuple representation"""
    rowsum = np.array(features.sum(1))
    r_inv = np.power(rowsum, -1).flatten()
    r_inv[np.isinf(r_inv)] = 0.
    r_mat_inv = sp.diags(r_inv)
    features = r_mat_inv.dot(features)
    return sparse_to_tuple(features)


def preprocess_reddit_features(features):
    from sklearn.preprocessing import StandardScaler
    scaler = StandardScaler()
    scaler.fit(features)
    features = scaler.transform(features)
    return features


def normalize_adj(adj):
    """Symmetrically normalize adjacency matrix."""
    adj = sp.coo_matrix(adj)
    rowsum = np.array(adj.sum(1))
    d_inv_sqrt = np.power(rowsum, -0.5).flatten()
    d_inv_sqrt[np.isinf(d_inv_sqrt)] = 0.
    d_mat_inv_sqrt = sp.diags(d_inv_sqrt)
    return adj.dot(d_mat_inv_sqrt).transpose().dot(d_mat_inv_sqrt).tocoo()


def preprocess_adj(adj, normalize=True):
    """Preprocessing of adjacency matrix for simple GCN model and conversion to tuple representation."""
    if normalize:
        adj_normalized = normalize_adj(adj + sp.eye(adj.shape[0]))
        return sparse_to_tuple(adj_normalized)
    else:
        return sparse_to_tuple(sp.coo_matrix(adj + sp.eye(adj.shape[0])))


def preprocess_adj_semiemb(adj):
    """Preprocessing of adjacency matrix for simple GCN model and conversion to tuple representation."""
    adj_normalized = sp.eye(adj.shape[0]) - normalize_adj(adj + sp.eye(adj.shape[0]))
    return sparse_to_tuple(adj_normalized)


def construct_feed_dict(features, support, labels, labels_mask, placeholders):
    """Construct feed dictionary."""
    feed_dict = dict()
    feed_dict.update({placeholders['labels']: labels})
    feed_dict.update({placeholders['labels_mask']: labels_mask})
    feed_dict.update({placeholders['features']: features})
    feed_dict.update({placeholders['support'][i]: support[i] for i in range(len(support))})
    feed_dict.update({placeholders['num_features_nonzero']: features[1].shape})
    return feed_dict


def chebyshev_polynomials(adj, k):
    """Calculate Chebyshev polynomials up to order k. Return a list of sparse matrices (tuple representation)."""
    print("Calculating Chebyshev polynomials up to order {}...".format(k))

    adj_normalized = normalize_adj(adj)
    laplacian = sp.eye(adj.shape[0]) - adj_normalized
    largest_eigval, _ = eigsh(laplacian, 1, which='LM')
    scaled_laplacian = (2. / largest_eigval[0]) * laplacian - sp.eye(adj.shape[0])

    t_k = list()
    t_k.append(sp.eye(adj.shape[0]))
    t_k.append(scaled_laplacian)

    def chebyshev_recurrence(t_k_minus_one, t_k_minus_two, scaled_lap):
        s_lap = sp.csr_matrix(scaled_lap, copy=True)
        return 2 * s_lap.dot(t_k_minus_one) - t_k_minus_two

    for i in range(2, k+1):
        t_k.append(chebyshev_recurrence(t_k[-1], t_k[-2], scaled_laplacian))

    return sparse_to_tuple(t_k)