def _supply_laplacians_etc_to_feed_dict(self, feed_dict, pairs, tvt):
if logging_enabled == True:
print(
"- Entered _supply_laplacians_etc_to_feed_dict Private Method")
for i, (g1, g2) in enumerate(pairs):
feed_dict[self._get_plhdr('features_1',
tvt)[i]] = g1.get_node_inputs()
feed_dict[self._get_plhdr('features_2',
tvt)[i]] = g2.get_node_inputs()
feed_dict[self._get_plhdr(
'num_nonzero_1', tvt)[i]] = g1.get_node_inputs_num_nonzero()
feed_dict[self._get_plhdr(
'num_nonzero_2', tvt)[i]] = g2.get_node_inputs_num_nonzero()
feed_dict[self._get_plhdr('edge_index_1', tvt)[i]] = g1.edge_index
feed_dict[self._get_plhdr('edge_index_2', tvt)[i]] = g2.edge_index
feed_dict[self._get_plhdr('incidence_mat_1',
tvt)[i]] = g1.incidence_mat
feed_dict[self._get_plhdr('incidence_mat_2',
tvt)[i]] = g2.incidence_mat
assert g1.incidence_mat is not None
num_laplacians = 1
for j in range(get_coarsen_level()):
for k in range(num_laplacians):
feed_dict[
self._get_plhdr('laplacians_1', tvt)[i][j][k]] = \
g1.get_laplacians(j)[k]
feed_dict[
self._get_plhdr('laplacians_2', tvt)[i][j][k]] = \
g2.get_laplacians(j)[k]
return feed_dict
def _supply_laplacians_etc_to_feed_dict(self, feed_dict, pairs, tvt):
if is_transductive():
gemb_lookup_ids_1 = []
gemb_lookup_ids_2 = []
for (g1, g2) in pairs:
gemb_lookup_ids_1.append(g1.global_id)
gemb_lookup_ids_2.append(g2.global_id)
feed_dict[self._get_plhdr('gemb_lookup_ids_1', tvt)] = \
gemb_lookup_ids_1
feed_dict[self._get_plhdr('gemb_lookup_ids_2', tvt)] = \
gemb_lookup_ids_2
else:
for i, (g1, g2) in enumerate(pairs):
feed_dict[self._get_plhdr('features_1', tvt)[i]] = \
g1.get_node_inputs()
feed_dict[self._get_plhdr('features_2', tvt)[i]] = \
g2.get_node_inputs()
feed_dict[self._get_plhdr('num_nonzero_1', tvt)[i]] = \
g1.get_node_inputs_num_nonzero()
feed_dict[self._get_plhdr('num_nonzero_2', tvt)[i]] = \
g2.get_node_inputs_num_nonzero()
num_laplacians = 1
for j in range(get_coarsen_level()):
for k in range(num_laplacians):
feed_dict[
self._get_plhdr('laplacians_1', tvt)[i][j][k]] = \
g1.get_laplacians(j)[k]
feed_dict[
self._get_plhdr('laplacians_2', tvt)[i][j][k]] = \
g2.get_laplacians(j)[k]
return feed_dict
def _coarsen(self, dense_node_inputs, adj):
assert ('metis_' in FLAGS.coarsening)
self.num_level = get_coarsen_level()
assert (self.num_level >= 1)
graphs, perm = coarsen(sp.csr_matrix(adj), levels=self.num_level,
self_connections=False)
permuted_padded_dense_node_inputs = perm_data(
dense_node_inputs.T, perm).T
self.sparse_permuted_padded_dense_node_inputs = self._preprocess_inputs(
sp.csr_matrix(permuted_padded_dense_node_inputs))
self.coarsened_laplacians = []
for g in graphs:
self.coarsened_laplacians.append([self._preprocess_adj(g.todense())])
assert (len(self.coarsened_laplacians) == self.num_laplacians * self.num_level + 1)
def _supply_laplacians_etc_to_feed_dict(self, feed_dict, origin_gs, pn_gs,
tvt):
num_laplacians = 1
for i, origin_g in enumerate(origin_gs):
feed_dict[self._get_plhdr('features_1', tvt)[i]] = \
origin_g.get_node_inputs()
feed_dict[self._get_plhdr('num_nonzero_1', tvt)[i]] = \
origin_g.get_node_inputs_num_nonzero()
for j in range(get_coarsen_level()):
for k in range(num_laplacians):
feed_dict[self._get_plhdr('laplacians_1', tvt)[i][j][k]] = \
origin_g.get_laplacians(j)[k]
for i, pn_g in enumerate(pn_gs):
feed_dict[self._get_plhdr('features_2', tvt)[i]] = \
pn_g.get_node_inputs()
feed_dict[self._get_plhdr('num_nonzero_2', tvt)[i]] = \
pn_g.get_node_inputs_num_nonzero()
for j in range(get_coarsen_level()):
for k in range(num_laplacians):
feed_dict[self._get_plhdr('laplacians_2', tvt)[i][j][k]] = \
pn_g.get_laplacians(j)[k]
return feed_dict
def _supply_laplacians_etc_to_feed_dict(self, feed_dict, pairs, tvt):
for i, (g1, g2) in enumerate(pairs):
feed_dict[self._get_plhdr('features_1', tvt)[i]] = \
g1.get_node_inputs()
feed_dict[self._get_plhdr('features_2', tvt)[i]] = \
g2.get_node_inputs()
feed_dict[self._get_plhdr('num_nonzero_1', tvt)[i]] = \
g1.get_node_inputs_num_nonzero()
feed_dict[self._get_plhdr('num_nonzero_2', tvt)[i]] = \
g2.get_node_inputs_num_nonzero()
num_laplacians = 1
for j in range(get_coarsen_level()):
for k in range(num_laplacians):
feed_dict[self._get_plhdr('laplacians_1', tvt)[i][j][k]] = \
g1.get_laplacians(j)[k]
feed_dict[self._get_plhdr('laplacians_2', tvt)[i][j][k]] = \
g2.get_laplacians(j)[k]
return feed_dict
def __init__(self, input_dim, data, dist_calculator):
self.input_dim = input_dim
self.batch_size = FLAGS.batch_size * (1 + FLAGS.num_neg)
# Create placeholders.
self.laplacians_1, self.laplacians_2, self.features_1, self.features_2, \
self.num_nonzero_1, self.num_nonzero_2, self.dropout, \
self.val_test_laplacians_1, self.val_test_laplacians_2, \
self.val_test_features_1, self.val_test_features_2, \
self.val_test_num_nonzero_1, self.val_test_num_nonzero_2 = \
self._create_basic_placeholders(FLAGS.batch_size, self.batch_size,
level=get_coarsen_level())
# self.pos_interact_score = []
# Build the model.
super(SiameseRankingModel, self).__init__()
self.origin_gs, self.pos_gs, self.neg_gs = self._load_pos_neg_train_pairs(
data, dist_calculator)
def __init__(self, input_dim, data, dist_calculator):
self.input_dim = input_dim
print('original_input_dim', self.input_dim)
self.laplacians_1, self.laplacians_2, self.features_1, self.features_2, \
self.num_nonzero_1, self.num_nonzero_2, self.dropout, \
self.val_test_laplacians_1, self.val_test_laplacians_2, \
self.val_test_features_1, self.val_test_features_2, \
self.val_test_num_nonzero_1, self.val_test_num_nonzero_2 = \
self._create_basic_placeholders(FLAGS.batch_size, FLAGS.batch_size,
level=get_coarsen_level())
self.train_y_true = tf.placeholder(tf.float32,
shape=(FLAGS.batch_size, 1))
self.val_test_y_true = tf.placeholder(tf.float32, shape=(1, 1))
# Build the model.
super(SiameseRegressionModel, self).__init__()
self.sim_kernel = create_sim_kernel(FLAGS.sim_kernel,
get_flags('yeta'),
get_flags('scale'))
self.train_triples = self._load_train_triples(data, dist_calculator)
def __init__(self, input_dim, data, dist_sim_calculator):
self.input_dim = input_dim
print('original_input_dim', self.input_dim)
if is_transductive():
self._create_transductive_gembs_placeholders(
data, FLAGS.batch_size, FLAGS.batch_size)
else:
self._create_basic_placeholders(FLAGS.batch_size,
FLAGS.batch_size,
level=get_coarsen_level())
self.train_y_true = tf.placeholder(tf.float32,
shape=(FLAGS.batch_size, 1))
self.val_test_y_true = tf.placeholder(tf.float32, shape=(1, 1))
# Build the model.
super(SiameseRegressionModel, self).__init__()
self.ds_kernel = create_ds_kernel(FLAGS.ds_kernel, get_flags('yeta'),
get_flags('scale'))
self.train_triples = self._load_train_triples(data,
dist_sim_calculator)
def __init__(self, input_dim, data, dist_calculator):
self.input_dim = input_dim
print('original_input_dim', self.input_dim)
self.num_class = 2
self.laplacians_1, self.laplacians_2, self.features_1, self.features_2, \
self.num_nonzero_1, self.num_nonzero_2, self.dropout, \
self.val_test_laplacians_1, self.val_test_laplacians_2, \
self.val_test_features_1, self.val_test_features_2, \
self.val_test_num_nonzero_1, self.val_test_num_nonzero_2 = \
self._create_basic_placeholders(FLAGS.batch_size, FLAGS.batch_size,
level=get_coarsen_level())
self.train_y_true = tf.placeholder(
tf.float32, shape=(FLAGS.batch_size, self.num_class))
self.val_test_y_true = tf.placeholder(
tf.float32, shape=(1, self.num_class))
# Build the model.
super(SiameseClassificationModel, self).__init__()
self.pos_pairs, self.neg_pairs = self._load_pos_neg_train_pairs(
data, dist_calculator)
self.cur_sample_class = 1 # 1 for pos, -1 for neg
def __init__(self, input_dim, data, dist_sim_calculator):
if logging_enabled == True:
print(
"- Entered SiameseRegressionModel::__init__ Constructor Method"
)
self.input_dim = input_dim
print('original_input_dim', self.input_dim)
self._create_basic_placeholders(ec.batch_size,
ec.batch_size,
level=get_coarsen_level())
self.train_y_true = tf.compat.v1.placeholder(tf.float32,
shape=(ec.batch_size, 1))
self.val_test_y_true = tf.compat.v1.placeholder(tf.float32,
shape=(1, 1))
# Build the model.
super(SiameseRegressionModel, self).__init__()
self.ds_kernel = create_ds_kernel(ec.ds_kernel, ec.yeta, ec.scale)
self.train_triples = self._load_train_triples(data,
dist_sim_calculator)
print("dist_sim_calculator-object_type", dist_sim_calculator)