def __init__(self, **kwargs):
# init some parameters
self.g_type = kwargs.get('graph_type', 'powerlaw')
#'erdos_renyi', 'powerlaw', 'small-world', 'barabasi_albert'
global NUM_MIN
global NUM_MAX
global MAX_ITERATION
global combineID
NUM_MIN = kwargs.get('NUM_MIN', NUM_MIN)
NUM_MAX = kwargs.get('NUM_MAX', NUM_MAX)
MAX_ITERATION = kwargs.get('MAX_ITERATION', MAX_ITERATION)
combineID = kwargs.get('combineID', combineID)
self.weighted = kwargs.get("weighted", False)
self.embedding_size = EMBEDDING_SIZE
self.learning_rate = LEARNING_RATE
self.reg_hidden = REG_HIDDEN
self.TrainSet = graph.py_GSet()
self.TestSet = graph.py_GSet()
self.utils = utils.py_Utils()
self.TrainBetwList = []
self.TestBetwList = []
self.metrics = metrics.py_Metrics()
self.inputs = dict()
self.activation = tf.nn.leaky_relu #leaky_relu relu selu elu
self.ngraph_train = 0
self.ngraph_test = 0
tf.compat.v1.disable_v2_behavior()
# [node_cnt, node_feat_dim]
self.node_feat = tf.compat.v1.placeholder(tf.float32, name="node_feat")
# [node_cnt, aux_feat_dim]
self.aux_feat = tf.compat.v1.placeholder(tf.float32, name="aux_feat")
# [node_cnt, node_cnt]
self.n2nsum_param = tf.compat.v1.sparse_placeholder(tf.float64, name="n2nsum_param")
# [node_cnt,1]
self.label = tf.compat.v1.placeholder(tf.float32, shape=[None,1], name="label")
# sample node pairs to compute the ranking loss
self.pair_ids_src = tf.compat.v1.placeholder(tf.int32, shape=[1,None], name='pair_ids_src')
self.pair_ids_tgt = tf.compat.v1.placeholder(tf.int32, shape=[1,None], name='pair_ids_tgt')
self.loss, self.trainStep, self.betw_pred, self.node_embedding, self.param_list = self.BuildNet()
self.saver = tf.compat.v1.train.Saver(max_to_keep=None)
config = tf.compat.v1.ConfigProto(device_count={"CPU": 8}, # limit to num_cpu_core CPU usage
inter_op_parallelism_threads=100,
intra_op_parallelism_threads=100,
log_device_placement=False)
config.gpu_options.allow_growth = True
self.session = tf.compat.v1.Session(config=config)
self.session.run(tf.compat.v1.global_variables_initializer())
import pickle as cp
import time
import fire
import networkx as nx
import tensorflow as tf
from tensorflow.keras.models import load_model, Model
from tqdm import trange, tqdm
import metrics
from betlearn import DataGenerator, EvaluateCallback
metrics = metrics.py_Metrics()
def evaluate_synthetic_data(data_test, model_path):
""" This function is most probably wrong because the synthetic data is one file per graph and score """
model = load_model(model_path, custom_objects={'tf': tf}, compile=False)
data = DataGenerator(tag='Synthetic', include_idx_map=True, random_samples=False, compute_betweenness=False)
evaluate = EvaluateCallback(data, prepend_str='')
evaluate.set_model(model)
with open(data_test, 'rb') as f:
valid_data = cp.load(f)
graph_list = valid_data[0]
betweenness_list = valid_data[1]
for i in trange(100):
g = graph_list[i]
data.add_graph(g)
data.betweenness = betweenness_list
def __init__(self, data_generator, prepend_str: str = 'val_'):
super().__init__()
self.data_generator = data_generator
self.prepend_str = prepend_str
self.metrics = metrics.py_Metrics()
self._supports_tf_logs = True