def plot_ground_truth():
map = np.array(list(range(100))).reshape([10, 10])
matrix = np.array([[i, j] for i in range(10) for j in range(10)])
plot_embeddings(map,
matrix,
file=FINAL_EMBEDDING_PATH + '/ground_truth.png')
def main(args):
nx_G = nx.read_edgelist(args.input, create_using=nx.DiGraph(), nodetype=None, data=[("weight", int)])
model = SDNE(nx_G, hidden_size=[512, 256, 256])
model.train(batch_size=3000, epochs=50, verbose=2)
embeddings = model.get_embeddings()
embeddings = {str(k): embeddings[k] for k in embeddings.keys()}
plot_embeddings(embeddings, args.label_file)
def main():
trainer = ParticleTrainer(embedding=USE_EMBEDDING, store_embeddings=PLOT_EMBEDDING_HISTORY)
trainer.train()
if USE_EMBEDDING and PLOT_EMBEDDING_HISTORY:
embedding_history = trainer.embedding_history
embedding_map = trainer.embedding_map.numpy()
for i, embedding in enumerate(embedding_history):
plot_embeddings(embedding_map, embedding, annotate=False, file=EMBEDDING_HISTORY_PATH + f'/{i:02d}.png')
def main(args):
# nx_G = read_graph()
nx_G = nx.read_edgelist(args.input,
create_using=nx.DiGraph(),
nodetype=int,
data=(("weight", int)))
line = Line(nx_G, batch_size=1024, embedding_size=128, order='second')
line.train(epoch=50, verbose=2)
_embeddings = line.get_embeddings()
_embeddings = {str(k): _embeddings[k] for k in _embeddings.keys()}
plot_embeddings(_embeddings, args.label_file)
def main(args):
nx_G = read_graph()
G = deepwalk(nx_G, args.directed, args.p, args.q)
walks = G.simulate_walks(args.num_walks, args.walk_length)
model = learning_walks(walks)
_embeddings = {}
for word in nx_G.nodes():
_embeddings[str(word)] = model.wv[str(word)]
plot_embeddings(_embeddings, args.label_file)
def main(args):
nx_G = read_graph()
G = node2vec_walk(nx_G, args.directed, args.p, args.q)
G.preprocess_transition_probs()
walks = G.simulate_walks(args.num_walks, args.walk_length)
model = learning_walks(walks)
_embeddings = {}
for v in nx_G.nodes():
_embeddings[str(v)] = model.wv[str(v)]
plot_embeddings(_embeddings, args.label_file)
def main():
trainer = ParticleTrainer(embedding=True)
trainer.load_snapshot(BEST_EMBEDDED_SNAPSHOT_PATH)
embedding_map = trainer.embedding_map.numpy()
embedding_matrix = trainer.net.embedding.weight
embedding_matrix = embedding_matrix.cpu().detach().numpy()
plot_ground_truth()
plot_embeddings(
embedding_map,
embedding_matrix,
file=FINAL_EMBEDDING_PATH +
f'/{basename(SNAPSHOT)}.png' if SAVE_FILE else None,
)
import numpy as np
import matplotlib.pyplot as plt
import networkx as nx
from sklearn.manifold import TSNE
from struc2vec import Struc2Vec
from utils import evaluate_embeddings, plot_embeddings
if __name__ == "__main__":
G = nx.read_edgelist('./data/Wiki_edgelist.txt', create_using=nx.DiGraph(), nodetype=None,
data=[('weight', int)])
model = Struc2Vec(G, 20, 80, workers=4, verbose=40, opt3_num_layers=5)
model.train(embed_size=256)
embeddings = model.get_embeddings()
evaluate_embeddings(embeddings, "./data/wiki_labels.txt")
plot_embeddings(embeddings, "./data/wiki_labels.txt")
feature_less=FEATURE_LESS, )
model.compile(optimizer=Adam(0.01), loss='categorical_crossentropy',
weighted_metrics=['categorical_crossentropy', 'acc'])
NB_EPOCH = 200
PATIENCE = 200 # early stopping patience
val_data = (model_input, y_val, val_mask)
mc_callback = ModelCheckpoint('./best_model.h5',
monitor='val_weighted_categorical_crossentropy',
save_best_only=True,
save_weights_only=True)
# train
print("start training")
model.fit(model_input, y_train, sample_weight=train_mask, validation_data=val_data,
batch_size=A.shape[0], epochs=NB_EPOCH, shuffle=False, verbose=2, callbacks=[mc_callback])
# test
model.load_weights('./best_model.h5')
eval_results = model.evaluate(
model_input, y_test, sample_weight=test_mask, batch_size=A.shape[0])
print('Done.\n'
'Test loss: {}\n'
'Test weighted_loss: {}\n'
'Test accuracy: {}'.format(*eval_results))
embedding_model = Model(model.input, outputs=Lambda(lambda x: model.layers[-1].output)(model.input))
embedding_weights = embedding_model.predict(model_input, batch_size=A.shape[0])
y = np.genfromtxt("{}{}.content".format('../data/cora/', 'cora'), dtype=np.dtype(str))[:, -1]
plot_embeddings(embedding_weights, np.arange(A.shape[0]), y)
words, word2ind, freqs = utils.extractDictionary(corpus, limit=20000)
data = utils.extractWordContextPairs(corpus, windowSize, word2ind)
del corpus
U0 = (np.random.rand(len(words), embDim) - 0.5) / embDim
V0 = (np.random.rand(len(words), embDim) - 0.5) / embDim
seq = sampling.createSamplingSequence(freqs)
contextFunction = lambda c: sampling.sampleContext(c, seq, negativesCount)
if len(sys.argv)>1 and sys.argv[1] == 'cumulative':
U,V = w2v_sgd.stochasticGradientDescend(data,np.copy(U0),np.copy(V0),contextFunction,grads.lossAndGradientCumulative)
else:
U,V = w2v_sgd.stochasticGradientDescend(data,np.copy(U0),np.copy(V0),contextFunction,grads.lossAndGradientBatched)
np.save('w2v-U',U)
np.save('w2v-V',V)
E = np.concatenate([U,V],axis=1)
E_reduced =utils.SVD_k_dim(E,k=2)
E_normalized_2d = E_reduced /np.linalg.norm(E_reduced, axis=1)[:, np.newaxis]
sampleWords = 'януари октомври седмица година медии пазар стоки бизнес фирма бюджет петрол нефт'.split()
utils.plot_embeddings(E_normalized_2d, word2ind, sampleWords, 'embeddings')
def hard_triplet_baseline_exp(device='3',
lr=1e-3,
n_epochs=300,
n_classes=10,
n_samples=12,
margin=0.3,
log_interval=50):
"""
:param device:
:param lr:
:param n_epochs:
:param n_classes:
:param n_samples:
:return:
"""
os.environ['CUDA_VISIBLE_DEVICES'] = str(device)
# get the mean and std of dataset train/a
standarizer = TaskbStandarizer(data_manager=Dcase18TaskbData())
mu, sigma = standarizer.load_mu_sigma(mode='train', device='a')
# get the normalized train dataset
train_dataset = DevSet(mode='train',
device='a',
transform=Compose(
[Normalize(mean=mu, std=sigma),
ToTensor()]))
test_dataset = DevSet(mode='test',
device='a',
transform=Compose(
[Normalize(mean=mu, std=sigma),
ToTensor()]))
train_batch_sampler = BalanceBatchSampler(dataset=train_dataset,
n_classes=n_classes,
n_samples=n_samples)
train_batch_loader = DataLoader(dataset=train_dataset,
batch_sampler=train_batch_sampler,
num_workers=1)
test_batch_sampler = BalanceBatchSampler(dataset=test_dataset,
n_classes=n_classes,
n_samples=n_samples)
test_batch_loader = DataLoader(dataset=test_dataset,
batch_sampler=test_batch_sampler,
num_workers=1)
model = networks.embedding_net_shallow()
model = model.cuda()
loss_fn = OnlineTripletLoss(
margin=margin,
triplet_selector=utils.RandomNegativeTripletSelector(margin=margin))
optimizer = optim.Adam(model.parameters(), lr=lr)
scheduler = lr_scheduler.StepLR(optimizer=optimizer,
step_size=30,
gamma=0.5)
fit(train_loader=train_batch_loader,
val_loader=test_batch_loader,
model=model,
loss_fn=loss_fn,
optimizer=optimizer,
scheduler=scheduler,
n_epochs=n_epochs,
log_interval=log_interval,
metrics=[AverageNoneZeroTripletsMetric()])
train_embedding_tl, train_labels_tl = utils.extract_embeddings(
train_batch_loader, model)
utils.plot_embeddings(train_embedding_tl, train_labels_tl)
test_embedding_tl, test_labels_tl = utils.extract_embeddings(
test_batch_loader, model)
utils.plot_embeddings(test_embedding_tl, test_labels_tl)
