def test_check_task_dataset_model_match():
variants = list(
gen_variants(dataset=["cora"], model=["gcn", "gat"], seed=[1, 2]))
variants.append(
namedtuple("Variant", ["dataset", "model", "seed"])(dataset="cora",
model="deepwalk",
seed=1))
variants = check_task_dataset_model_match("node_classification", variants)
assert len(variants) == 4
def test_gen_variants():
variants = list(
gen_variants(dataset=["cora"], model=["gcn", "gat"], seed=[1, 2]))
assert len(variants) == 4
mp.set_start_method("spawn", force=True)
parser = options.get_training_parser()
args, _ = parser.parse_known_args()
args = options.parse_args_and_arch(parser, args)
# Make sure datasets are downloaded first
datasets = args.dataset
for dataset in datasets:
args.dataset = dataset
_ = build_dataset(args)
args.dataset = datasets
print(args)
variants = list(
gen_variants(dataset=args.dataset, model=args.model, seed=args.seed))
device_ids = args.device_id
if args.cpu:
num_workers = 1
else:
num_workers = len(device_ids)
print("num_workers", num_workers)
results_dict = defaultdict(list)
with mp.Pool(processes=num_workers) as pool:
# Map process to cuda device
pids = pool.map(getpid, range(num_workers))
pid_to_cuda = dict(zip(pids, device_ids))
# yield all variants