def prepare_data_for_stellargraph(self):
def load_raw_input():
adj = np.load(os.path.join(self.data_path, 'adj.pkl'),
allow_pickle=True)
features = np.load(os.path.join(self.data_path, 'features.pkl'),
allow_pickle=True)
labels = np.load(os.path.join(self.data_path, 'train.pkl'),
allow_pickle=True)
return adj, features, labels
print("Reading raw inputs...")
adj, features, labels = load_raw_input()
print("creating nodes...")
adj_list = [[i, j, adj[i, j]] for i, j in zip(*adj.nonzero())]
tmp_df = pd.DataFrame(adj_list)
tmp_df.columns = ["source", "target", "weight"]
print("creating edges...")
feature_df = pd.DataFrame(features)
feature_df.columns = [f"w{i}" for i in range(feature_df.shape[1])]
print("creating labels...")
label_series = pd.DataFrame({"label": labels})["label"]
my_graph = StellarGraph({"paper": feature_df}, {"cites": tmp_df})
print(my_graph.info())
return my_graph, label_series
def load_from_file(filePrefix):
nodes_filename = filePrefix + "_nodes.txt"
edges_filename = filePrefix + "_edges.txt"
node_features = None
edge_features = None
#https://stellargraph.readthedocs.io/en/stable/demos/basics/loading-numpy.html
with open(nodes_filename) as f:
num_nodes, num_node_features = map(int, f.readline().split('\t')[:-1])
if num_node_features > 0:
node_features = np.zeros((num_node_features, num_nodes))
for i, line in enumerate(f.readlines()):
features = np.array(list(map(float, line.split('\t')[1:-1])))
for fIndex in range(num_node_features):
node_features[fIndex][i] = features[fIndex]
#node_features[i] = features
# read edge features
with open(edges_filename) as f:
num_edges, num_edge_features = map(int, f.readline().split('\t')[:-1])
senders = np.zeros(num_edges, dtype=int)
receivers = np.zeros(num_edges, dtype=int)
if num_edge_features > 0:
edge_features = np.zeros((num_edge_features, num_edges))
for i, line in enumerate(f.readlines()):
elements = line.split('\t')
senders[i] = int(elements[0])
receivers[i] = int(elements[1])
if edge_features is not None:
features = np.array(list(map(float, elements[2:-1])))
for fIndex in range(num_edge_features):
edge_features[fIndex][i] = features[fIndex]
#edge_features[i] = np.array(list(map(float, elements[2:-1])))
square_numeric_edges = pd.DataFrame({
"source": senders,
"target": receivers
})
square_node_data = pd.DataFrame({"x": node_features[0].tolist()})
#square_node_data = pd.DataFrame( { "x": node_features[0].tolist(), "y": node_features[1].tolist(), "z" : node_features[2].tolist() } )
#feature_array = np.array([[1.0, -0.2], [2.0, 0.3], [3.0, 0.0], [4.0, -0.5]], dtype=np.float32)
#print("node_features")
#print(node_features)
#print("square_numeric_edges")
#print(square_numeric_edges)
square_numeric = StellarGraph(square_node_data, edges=square_numeric_edges)
print("GRAPH INFO")
print("....................................")
print(square_numeric.info())
print("....................................")
return square_numeric
import utils_ctdne as utils
from stellargraph import StellarGraph
from stellargraph.data import EdgeSplitter
from sklearn.model_selection import train_test_split
# import data and create initial graph G
# filename_edges = "Datasets/Cora/cora_cites.csv"
# wt = False
filename_edges = "Datasets/soc-sign-bitcoinotc/soc-sign-bitcoinotc-temporal.csv"
wt=True
edgelist_df = utils.load_data(filename_edges, weighted = wt)
G = StellarGraph(edges = edgelist_df)
print("\n", G.info())
print("Created master graph from data")
# Define an edge splitter on the original graph:
edge_splitter_test = EdgeSplitter(G)
# Randomly sample a fraction p=0.1 of all positive links,
# and same number of negative links, from graph, and obtain the
# reduced graph graph_test with the sampled links removed:
(
G_test, # To compute node embeddings with mode edges than G_train
examples_test,
labels_test
) = edge_splitter_test.train_test_split(p=0.1,
method="global")