def evaluate_sage():
fname = iterate_datasets()
os.chdir("/nfs/zty/Graph/Dynamic-Graph")
cmd = "python -m experiment.graphsage --dataset {dataset} --epochs 50 --dropout 0.2 --weight_decay 1e-5 --learning_rate=0.0001 --nodisplay "
commands = [cmd.format(dataset=name) for name in fname]
print("Preprocessing finished.")
Parallel(n_jobs=args.n_jobs)(delayed(os.system)(cmd) for cmd in commands)
def evaluate_htne(project_dir="/nfs/zty/Graph/4-htne/emb"):
fname = iterate_datasets(dataset=args.dataset)
fname = fname[args.start:args.end]
if args.run:
logger.info("Running {} embedding programs.".format(args.method))
run_htne(dataset=args.dataset, n_jobs=args.n_jobs, fname=fname)
logger.info("Done training embedding.")
else:
logger.info("Use pretrained {} embeddings.".format(args.method))
for name in fname:
logger.info(name)
edgel, nodel = load_label_edges(dataset=name)
train_edges, valid_edges, test_edges = id_map(edgel[0], nodel[0])
for hist_len in [20]:
fpath = "{}/{}.emb{}".format(project_dir, name, hist_len)
id2idx, embeds = load_embeddings(fpath, skiprows=1, sep=" ")
X_train = edge2tabular(train_edges, id2idx, embeds)
y_train = train_edges["label"]
X_valid = edge2tabular(valid_edges, id2idx, embeds)
y_valid = valid_edges["label"]
X_test = edge2tabular(test_edges, id2idx, embeds)
y_test = test_edges["label"]
# print(X_train.shape, y_train.shape, X_test.shape, y_test.shape)
vauc, acc, f1, auc = lr_evaluate(X_train, y_train, X_valid,
y_valid, X_test, y_test)
write_result(name, "htne", {
"hist_len": hist_len,
"epoch": 50
}, (vauc, acc, f1, auc))
def evaluate_node2vec(project_dir="/nfs/zty/Graph/0-node2vec/emb"):
fname = iterate_datasets()
fname = fname[args.start:args.end]
if args.run:
logger.info("Running {} embedding programs.".format("node2vec"))
run_node2vec(dataset=args.dataset,
n_jobs=args.n_jobs,
fname=fname,
start=args.start,
end=args.end,
times=args.times)
logger.info("Done node2vec embedding.")
else:
logger.info("Use pretrained {} embeddings.".format("node2vec"))
for name, p, q in product(fname, [0.25, 0.5, 1, 2, 4],
[0.25, 0.5, 1, 2, 4]):
logger.info("dataset={}, p={:.2f}, q={:.2f}".format(name, p, q))
edges, nodes = load_label_edges(dataset=name)
train_edges, valid_edges, test_edges = id_map(edges[0], nodes[0])
fpath = "{}/{}-{p:.2f}-{q:.2f}.emb".format(project_dir, name, p=p, q=q)
id2idx, embs = load_embeddings(fpath, skiprows=1)
X_train = edge2tabular(train_edges, id2idx, embs)
y_train = train_edges["label"]
X_valid = edge2tabular(valid_edges, id2idx, embs)
y_valid = valid_edges["label"]
X_test = edge2tabular(test_edges, id2idx, embs)
y_test = test_edges["label"]
# print(X_train.shape, y_train.shape, X_test.shape, y_test.shape)
vauc, acc, f1, auc = lr_evaluate(X_train, y_train, X_valid, y_valid,
X_test, y_test)
write_result(name, "node2vec", {"p": p, "q": q}, (vauc, acc, f1, auc))
pass
def evaluate_ctdne(project_dir="/nfs/zty/Graph/Dynamic-Graph/ctdne_embs"):
fname = iterate_datasets(dataset=args.dataset)
fname = fname[args.start:args.end]
if args.run:
logger.info("Running {} embedding programs.".format(args.method))
Parallel(n_jobs=args.n_jobs)(delayed(run_ctdne)(fname=[name])
for name in fname)
logger.info("Done {} embeddings.".format(args.method))
for name in fname:
logger.info(
"dataset={}, num_walk=10, walk_length=80, context_window=10".
format(name))
fpath = "{}/{}.emb".format(project_dir, name)
id2idx, embeds = load_embeddings(fpath, skiprows=0, sep=" ")
edgel, nodel = load_label_edges(dataset=name)
train_edges, valid_edges, test_edges = id_map(edgel[0], nodel[0])
X_train = edge2tabular(train_edges, id2idx, embeds)
y_train = train_edges["label"]
X_valid = edge2tabular(valid_edges, id2idx, embeds)
y_valid = valid_edges["label"]
X_test = edge2tabular(test_edges, id2idx, embeds)
y_test = test_edges["label"]
# print(X_train.shape, y_train.shape, X_test.shape, y_test.shape)
vauc, acc, f1, auc = lr_evaluate(X_train, y_train, X_valid, y_valid,
X_test, y_test)
write_result(name, "ctdne", {
"num_walk": 10,
"walk_length": 80,
"context_window": 10
}, (vauc, acc, f1, auc))
def run_htne(dataset="all",
project_dir="/nfs/zty/Graph/4-htne/",
n_jobs=4,
**kwargs):
fname = iterate_datasets(dataset=dataset)
command = "python {project_dir}/HTNE.py -d {input_path} -o {output_path} --hist-len {hist_len}"
commands = []
for name in fname:
input_path = os.path.join(project_dir, "data/{}.edges".format(name))
if not os.path.exists(input_path):
df, nodes = load_data(dataset=name, mode="train")[0]
df["timestamp"] = (df["timestamp"] - df["timestamp"].min()) / \
(df["timestamp"].max() - df["timestamp"].min())
id2idx = {row.node_id: row.id_map for row in nodes.itertuples()}
df["from_node_id"] = df["from_node_id"].map(id2idx)
df["to_node_id"] = df["to_node_id"].map(id2idx)
df = df[["from_node_id", "to_node_id", "timestamp"]]
df.to_csv(input_path, index=None, header=None, sep=" ")
output_path = os.path.join(project_dir, "emb/{}.emb".format(name))
for hist_len in [20]:
hist_path = output_path + str(hist_len)
commands.append(
command.format(project_dir=project_dir,
input_path=input_path,
output_path=hist_path,
hist_len=hist_len))
print("Preprocessing finished.")
Parallel(n_jobs=n_jobs)(delayed(os.system)(cmd) for cmd in commands)
def evaluate_tnode():
fname = iterate_datasets(dataset=args.dataset)
fname = fname[args.start:args.end]
logger.info("Running {} embedding programs.".format(args.method))
run_tnode(dataset=args.dataset,
n_jobs=args.n_jobs,
start=args.start,
end=args.end)
logger.info("Done {}.".format(args.method))
def evaluate_tgat(project_dir="/nfs/zty/Graph/TGAT-bk"):
fname = iterate_datasets(dataset=args.dataset)
fname = fname[args.start:args.end]
# command = "python {}/exper_edge.py -d {} --gpu {} -f --uniform "
# command = "python {}/exper_edge.py -d {} --gpu {} -f"
# command = "python {}/exper_edge.py -d {} --gpu {} --uniform"
command = "python {}/exper_edge.py -d {} --gpu {} --time empty "
commands = []
for name in fname:
commands.append(command.format(project_dir, name, args.gid))
os.chdir(project_dir)
print("Preprocessing finished.")
for cmd in commands:
os.system(cmd)
def run_tnode(dataset="all",
project_dir="/nfs/zty/Graph/5-tNodeEmbed/",
n_jobs=16,
**kwargs):
fname = iterate_datasets(dataset=dataset)
fname = fname[kwargs["start"]:kwargs["end"]]
command = "python {project_dir}/src/main.py -d {dataset} -n {nstep}"
commands = []
for name, nstep in product(fname, [128, 32, 8]):
dump_foler = os.path.join(project_dir, "data/{}".format(name))
if not os.path.exists(dump_foler):
os.makedirs(dump_foler)
os.chmod(dump_foler, 0o777)
cmd = command.format(project_dir=project_dir,
dataset=name,
nstep=nstep)
commands.append(cmd)
os.chdir(project_dir)
print("Preprocessing finished.")
Parallel(n_jobs=n_jobs)(delayed(os.system)(cmd) for cmd in commands)
def evaluate_triad(project_dir="/nfs/zty/Graph/2-DynamicTriad/output"):
fname = iterate_datasets(dataset=args.dataset)
fname = fname[args.start:args.end]
if args.run:
logger.info("Running {} embedding programs.".format(args.method))
run_triad(dataset=args.dataset,
n_jobs=args.n_jobs,
fname=fname,
start=args.start,
end=args.end,
times=args.times)
logger.info("Done training embedding.")
else:
logger.info("Use pretrained {} embeddings.".format(args.method))
for name, stepsize in product(fname, [1, 4, 8]):
logger.info(name)
edgel, nodel = load_label_edges(dataset=name)
train_edges, valid_edges, test_edges = id_map(edgel[0], nodel[0])
fdir = "{}/{}-{}/".format(project_dir, name, stepsize)
step_embeds = [
load_embeddings(fdir + f, skiprows=0) for f in os.listdir(fdir)
]
id2idx, embeds = step_embeds[-1]
X_train = edge2tabular(train_edges, id2idx, embeds)
y_train = train_edges["label"]
X_valid = edge2tabular(valid_edges, id2idx, embeds)
y_valid = valid_edges["label"]
X_test = edge2tabular(test_edges, id2idx, embeds)
y_test = test_edges["label"]
# print(X_train.shape, y_train.shape, X_test.shape, y_test.shape)
vauc, acc, f1, auc = lr_evaluate(X_train, y_train, X_valid, y_valid,
X_test, y_test)
write_result(name, "triad", {
"beta_1": 0.1,
"beta_2": 0.1,
"stepsize": stepsize
}, (vauc, acc, f1, auc))
def run_gta(dataset="all",
project_dir="/nfs/zty/Graph/Dynamic-Graph/",
n_jobs=4,
**kwargs):
start, end = kwargs["start"], kwargs["end"]
fname = iterate_datasets(dataset=dataset)[start:end]
os.chdir(project_dir)
command = "python main.py --dataset {dataset} --epochs 50 --dropout 0.2 --weight_decay 1e-5 --learning_rate=0.0001 --nodisplay "
commands = []
comps = []
for name in fname:
cmd = command.format(dataset=name)
commands.append(cmd)
commands.append(cmd + " --nodynamic_neighbor")
commands.append(cmd + " --sampler temporal")
commands.append(cmd +
" --sampler temporal --use_context --context_size 20")
commands.append(cmd +
" --sampler temporal --use_context --context_size 40")
commands.append(cmd +
" --sampler temporal --use_context --context_size 80")
commands.append(cmd +
" --sampler temporal --use_context --context_size 160")
commands.append(cmd +
" --sampler mask --use_context --context_size 20")
commands.append(cmd +
" --sampler mask --use_context --context_size 40")
commands.append(cmd +
" --sampler mask --use_context --context_size 80")
commands.append(cmd +
" --sampler mask --use_context --context_size 160")
# comps = repeat_string(comps)
# commands = repeat_string(commands, times=1)
print("Preprocessing finished.")
# Parallel(n_jobs=n_jobs)(delayed(os.system)(cmd) for cmd in comps)
Parallel(n_jobs=n_jobs)(delayed(os.system)(cmd) for cmd in commands)