def test_graph_cross_validation():
dataset = build_dataset_from_name('imdb-b')
# first level, 10 folds
utils.graph_cross_validation(dataset, 10)
# set to fold id
utils.set_fold(dataset, 1)
# get train split
train_dataset = utils.graph_get_split(dataset, "train", False)
# further split train to train / val
utils.graph_random_splits(train_dataset, 0.8, 0.2)
import sys
sys.path.append('../')
from autogl.datasets import build_dataset_from_name, utils
from autogl.solver import AutoGraphClassifier
from autogl.module import Acc, BaseModel
dataset = build_dataset_from_name('mutag')
utils.graph_random_splits(dataset, train_ratio=0.4, val_ratio=0.4)
autoClassifier = AutoGraphClassifier.from_config(
'../configs/graph_classification.yaml')
# train
autoClassifier.fit(
dataset,
time_limit=3600,
train_split=0.8,
val_split=0.1,
cross_validation=True,
cv_split=10,
)
autoClassifier.get_leaderboard().show()
print('best single model:\n',
autoClassifier.get_leaderboard().get_best_model(0))
# test
predict_result = autoClassifier.predict_proba()
print(
Acc.evaluate(predict_result,
dataset.data.y[dataset.test_index].cpu().detach().numpy()))
deg_max = int(
degree(data.edge_index[0], data.num_nodes).max().item())
max_degree = max(max_degree, deg_max)
else:
max_degree = 0
for data, _ in _converted_dataset:
deg_max = data.in_degrees().max().item()
max_degree = max(max_degree, deg_max)
dataset = OneHotDegreeGenerator(max_degree).fit_transform(
dataset, inplace=False)
elif args.dataset == "collab":
from autogl.module.feature._auto_feature import OnlyConstFeature
dataset = OnlyConstFeature().fit_transform(dataset, inplace=False)
utils.graph_random_splits(dataset,
train_ratio=0.8,
val_ratio=0.1,
seed=args.seed)
autoClassifier = AutoGraphClassifier.from_config(args.configs)
# train
autoClassifier.fit(dataset, evaluation_method=[Acc], seed=args.seed)
autoClassifier.get_leaderboard().show()
print("best single model:\n",
autoClassifier.get_leaderboard().get_best_model(0))
# test
acc = autoClassifier.evaluate(metric="acc")
print("test acc {:.4f}".format(acc))
from autogl.datasets import build_dataset_from_name
from autogl.solver import AutoGraphClassifier
from autogl.datasets import utils
mutag = build_dataset_from_name("mutag")
utils.graph_random_splits(mutag, 0.8, 0.1)
solver = AutoGraphClassifier(graph_models=("gin", ),
hpo_module=None,
device="auto")
solver.fit(mutag, evaluation_method=["acc"])
result = solver.predict(mutag)
print(
"Acc:",
sum([d.data["y"].item() == r
for d, r in zip(mutag.test_split, result)]) / len(result))