util.py

import numpy as np
import pickle as pkl
import torch
import networkx as nx
import scipy.sparse as sp
import sys

exc_path = sys.path[0]

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


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(exc_path, 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(exc_path, 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

    features = sp.vstack((allx, tx)).tolil()
    features[test_idx_reorder, :] = features[test_idx_range, :]
    adj = nx.adjacency_matrix(nx.from_dict_of_lists(graph))
    adj = adj + adj.T.multiply(adj.T > adj) - adj.multiply(adj.T > adj)
    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)

    return adj, features, idx_train, idx_val, idx_test, labels


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 normalize_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 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 accuracy(output, labels):
    preds = output.max(1)[1].type_as(labels)
    correct = preds.eq(labels).double()
    correct = correct.sum()
    return correct / len(labels)


def sparse_mx_to_torch_sparse_tensor(sparse_mx):
    """Convert a scipy sparse matrix to a torch sparse tensor."""
    sparse_mx = sparse_mx.tocoo().astype(np.float32)
    indices = torch.from_numpy(
        np.vstack((sparse_mx.row, sparse_mx.col)).astype(np.int64))
    values = torch.from_numpy(sparse_mx.data)
    shape = torch.Size(sparse_mx.shape)
    return torch.sparse.FloatTensor(indices, values, shape)

from numpy.testing import assert_array_almost_equal

def build_uniform_P(size, noise):
    """ The noise matrix flips any class to any other with probability
    noise / (#class - 1).
    """

    assert (noise >= 0.) and (noise <= 1.)

    P = np.float64(noise) / np.float64(size - 1) * np.ones((size, size))
    np.fill_diagonal(P, (np.float64(1) - np.float64(noise)) * np.ones(size))

    diag_idx = np.arange(size)
    P[diag_idx, diag_idx] = P[diag_idx, diag_idx] + 1.0 - P.sum(0)
    assert_array_almost_equal(P.sum(axis=1), 1, 1)
    return P


def build_pair_p(size, noise):
    assert (noise >= 0.) and (noise <= 1.)
    P = (1.0 - np.float64(noise)) * np.eye(size)
    for i in range(size):
        P[i, i - 1] = np.float64(noise)
    assert_array_almost_equal(P.sum(axis=1), 1, 1)
    return P


def multiclass_noisify(y, P, random_state=0):
    """ Flip classes according to transition probability matrix T.
    It expects a number between 0 and the number of classes - 1.
    """

    assert P.shape[0] == P.shape[1]
    assert np.max(y) < P.shape[0]

    # row stochastic matrix
    assert_array_almost_equal(P.sum(axis=1), np.ones(P.shape[1]))
    assert (P >= 0.0).all()

    m = y.shape[0]
    new_y = y.copy()
    flipper = np.random.RandomState(random_state)

    for idx in np.arange(m):
        i = y[idx]
        # draw a vector with only an 1
        flipped = flipper.multinomial(1, P[i, :], 1)[0]
        new_y[idx] = np.where(flipped == 1)[0]

    return new_y

def noisify_with_P(y_train, nb_classes, noise, random_state=None,  noise_type='uniform'):

    if noise > 0.0:
        if noise_type=='uniform':
            print('Uniform noise')
            P = build_uniform_P(nb_classes, noise)
        elif noise_type == 'pair':
            print('Pair noise')
            P = build_pair_p(nb_classes, noise)
        else:
            print('Noise type have implemented')
        # seed the random numbers with #run
        y_train_noisy = multiclass_noisify(y_train, P=P,
                                           random_state=random_state)
        actual_noise = (y_train_noisy != y_train).mean()
        assert actual_noise > 0.0
        print('Actual noise %.2f' % actual_noise)
        y_train = y_train_noisy
    else:
        P = np.eye(nb_classes)

    return y_train, P