def test_graph_scnn():
from sklearn.metrics import accuracy_score
A, X, Y = data.parse_nci()
n_graphs = len(A)
indices = np.arange(n_graphs)
np.random.shuffle(indices)
train_indices = indices[:n_graphs // 3]
valid_indices = indices[n_graphs // 3:(2 * n_graphs) // 3]
test_indices = indices[(2 * n_graphs) // 3:]
scnn = GraphSCNN(n_hops=5)
scnn.fit(A,
X,
Y,
train_indices=train_indices,
valid_indices=valid_indices,
learning_rate=0.05,
batch_size=100,
n_epochs=500)
preds = scnn.predict(A, X, test_indices)
actuals = np.argmax(Y[test_indices, :], axis=1)
print 'Accuracy: %.4f' % (accuracy_score(actuals, preds))
def test_graph_scnn():
from sklearn.metrics import accuracy_score
A, X, Y = data.parse_nci()
n_graphs = len(A)
indices = np.arange(n_graphs)
np.random.shuffle(indices)
train_indices = indices[:n_graphs // 3]
valid_indices = indices[n_graphs // 3:(2* n_graphs) // 3]
test_indices = indices[(2* n_graphs) // 3:]
scnn = GraphSCNN(n_hops=5)
scnn.fit(A, X, Y, train_indices=train_indices, valid_indices=valid_indices,
learning_rate=0.05, batch_size=100, n_epochs=500)
preds = scnn.predict(A, X, test_indices)
actuals = np.argmax(Y[test_indices,:], axis=1)
print 'Accuracy: %.4f' % (accuracy_score(actuals, preds))
f1_micro = f1_score(actuals, preds, average='micro')
f1_macro = f1_score(actuals, preds, average='macro')
print 'form: name,micro_f,macro_f,accuracy'
print '###RESULTS###: %s,%s,%.6f,%.8f,%.8f,%.8f' % (
data_name, model_name, train_prop, f1_micro, f1_macro, accuracy)
if __name__ == '__main__':
np.random.seed()
args = sys.argv[1:]
name_to_data = {
'nci1':
lambda: data.parse_nci(graph_name='nci1.graph'),
'nci109':
lambda: data.parse_nci(graph_name='nci109.graph'),
'mutag':
lambda: data.parse_nci(graph_name='mutag.graph'),
'ptc':
lambda: data.parse_nci(graph_name='ptc.graph'),
'enzymes':
lambda: data.parse_nci(graph_name='enzymes.graph'),
'nci1struct':
lambda: data.parse_nci(graph_name='nci1.graph',
with_structural_features=True),
'nci109struct':
lambda: data.parse_nci(graph_name='nci109.graph',
with_structural_features=True),
}
f1_micro = f1_score(actuals, preds, average='micro')
f1_macro = f1_score(actuals, preds, average='macro')
accuracy = accuracy_score(actuals, preds)
print 'form: name,n_hops,transform_name,micro_f,macro_f,accuracy'
print '###RESULTS###: %s,%d,%s,%.8f,%.8f,%.8f' % (
name, n_hops, transform_name, f1_micro, f1_macro, accuracy)
if __name__ == '__main__':
np.random.seed()
args = sys.argv[1:]
if len(args) == 0:
graph_experiment(lambda: data.parse_nci('nci1.graph'), 'cora', 2)
else:
name_to_data = {
'nci1':
lambda: data.parse_nci('nci1.graph'),
#'nci1': lambda : data.parse_graph_data('nci1'),
'nci109':
lambda: data.parse_nci('nci109.graph'),
#'mutag': lambda : data.parse_graph_data('mutag'),
'mutag':
lambda: data.parse_nci('mutag.graph'),
'enzymes':
lambda: data.parse_nci('enzymes.graph'),
'ptc':
lambda: data.parse_nci('ptc.graph'),
'nci1struct':
actuals = Y[test_indices, :]
accuracy = accuracy_score(actuals, preds)
f1_micro = f1_score(actuals, preds, average="micro")
f1_macro = f1_score(actuals, preds, average="macro")
print "form: name,micro_f,macro_f,accuracy"
print "###RESULTS###: %s,%s,%.8f,%.8f,%.8f" % (data_name, model_name, f1_micro, f1_macro, accuracy)
if __name__ == "__main__":
np.random.seed()
args = sys.argv[1:]
name_to_data = {
"nci1": lambda: data.parse_nci(graph_name="nci1.graph"),
"nci109": lambda: data.parse_nci(graph_name="nci109.graph"),
"nci1struct": lambda: data.parse_nci(graph_name="nci1.graph", with_structural_features=True),
"nci109struct": lambda: data.parse_nci(graph_name="nci109.graph", with_structural_features=True),
}
name_to_parameters = {
"nci1": {
"num_dimensions": 2,
"kernel_type": 1,
"feature_type": 3,
"ds_name": "nci1",
"window_size": 2,
"ngram_type": 0,
"sampling_type": 1,
"graphlet_size": 0,
actuals = np.argmax(Y[test_indices,:], axis=1)
f1_micro = f1_score(actuals, preds, average='micro')
f1_macro = f1_score(actuals, preds, average='macro')
accuracy = accuracy_score(actuals, preds)
print 'form: name,n_hops,transform_name,micro_f,macro_f,accuracy'
print '###RESULTS###: %s,%d,%s,%.8f,%.8f,%.8f' % (name, n_hops, transform_name, f1_micro, f1_macro, accuracy)
if __name__ == '__main__':
np.random.seed()
args = sys.argv[1:]
if len(args) == 0:
graph_experiment(lambda : data.parse_nci('nci1.graph'), 'cora', 2)
else:
name_to_data = {
'nci1': lambda : data.parse_nci('nci1.graph'),
'nci109': lambda : data.parse_nci('nci109.graph'),
'nci1struct': lambda: data.parse_nci(graph_name='nci1.graph', with_structural_features=True),
'nci109struct': lambda: data.parse_nci(graph_name='nci109.graph', with_structural_features=True),
}
transform_lookup = {
'id': None,
'rwl': util.rw_laplacian,
'l': util.laplacian,
}
name = args[0]