def main():
args = parse_arguments()
config = get_config(args.config_file, is_test=args.test)
np.random.seed(config.seed)
torch.manual_seed(config.seed)
torch.cuda.manual_seed_all(config.seed)
config.use_gpu = config.use_gpu and torch.cuda.is_available()
# log info
log_file = os.path.join(config.save_dir, "log_exp_{}.txt".format(config.run_id))
logger = setup_logging(args.log_level, log_file)
logger.info("Writing log file to {}".format(log_file))
logger.info("Exp instance id = {}".format(config.run_id))
logger.info("Exp comment = {}".format(args.comment))
logger.info("Config =")
print(">" * 80)
pprint(config)
print("<" * 80)
# Run the experiment
try:
runner = eval(config.runner)(config)
if not args.test:
runner.train()
else:
runner.test()
except:
logger.error(traceback.format_exc())
sys.exit(0)
from utils.data_helper import create_graphs
import torch
from utils.arg_helper import get_config
from torch.utils.data import DataLoader
from random import shuffle
from torch.optim import Adam
from torch.optim.lr_scheduler import ReduceLROnPlateau
from classifier.losses import MulticlassClassificationLoss
class Bunch:
def __init__(self, **kwds):
self.__dict__.update(kwds)
config = get_config('config/gran_PROTEINS.yaml', is_test='false')
config.use_gpu = config.use_gpu and torch.cuda.is_available()
def data_to_bunch(data):
# for k in data[0].keys():
# l = data[0][k]
# if hasattr(l, 'shape'):
# print(k, " ", l.shape)
# else:
# print(k, " ", l)
num_nodes = data[0]['num_nodes_gt']
print('number of parameters : {}'.format(
sum([np.prod(x.shape) for x in model.parameters()])))
if not args.test:
trained_model = train_graph_generation(args, config, train_loader,
val_loader, test_loader, model)
else:
test_model = test(args, config, model, dataset)
if __name__ == '__main__':
"""
Process command-line arguments, then call main()
"""
args = parse_arguments()
config = get_config(args.config_file, is_test=args.test)
p_name = utils.project_name(config.dataset.name)
if args.wandb:
os.environ['WANDB_API_KEY'] = args.wandb_apikey
wandb.init(project='{}'.format(p_name),
name='{}-{}'.format(args.namestr, args.model_name))
''' Fix Random Seed '''
seed_everything(args.seed)
# Check if settings file
if os.path.isfile("settings.json"):
with open('settings.json') as f:
data = json.load(f)
args.wandb_apikey = data.get("wandbapikey")
main(args, config)
)
optimizer = optim.Adam(model.parameters(),
lr=config.train.lr_init,
betas=(0.9, 0.999),
eps=1e-8,
weight_decay=config.train.weight_decay)
fit(model,
optimizer,
mc_sampler,
train_dl,
max_node_number=config.dataset.max_node_num,
max_epoch=config.train.max_epoch,
config=config,
save_interval=config.train.save_interval,
sample_interval=config.train.sample_interval,
sigma_list=config.train.sigmas,
sample_from_sigma_delta=0.0,
test_dl=test_dl)
sample_main(config, args)
if __name__ == "__main__":
# torch.autograd.set_detect_anomaly(True)
args = parse_arguments('train_ego_small.yaml')
ori_config_dict = get_config(args)
config_dict = edict(ori_config_dict.copy())
process_config(config_dict)
print(config_dict)
train_main(config_dict, args)