val_score = run_an_eval_epoch(args, model, val_loader)
early_stop = stopper.step(val_score, model)
print('epoch {:d}/{:d}, validation {} {:.4f}, best validation {} {:.4f}'.format(
epoch + 1, args['num_epochs'], args['metric_name'], val_score,
args['metric_name'], stopper.best_score))
if early_stop:
break
if not args['pre_trained']:
stopper.load_checkpoint(model)
test_score = run_an_eval_epoch(args, model, test_loader)
print('test {} {:.4f}'.format(args['metric_name'], test_score))
if __name__ == "__main__":
import argparse
from configure import get_exp_configure
parser = argparse.ArgumentParser(description='Aromaticity Prediction')
parser.add_argument('-m', '--model', type=str, choices=['AttentiveFP'],
default='AttentiveFP', help='Model to use')
parser.add_argument('-p', '--pre-trained', action='store_true',
help='Whether to skip training and use a pre-trained model')
args = parser.parse_args().__dict__
args['dataset'] = 'Aromaticity'
args['exp'] = '_'.join([args['model'], args['dataset']])
args.update(get_exp_configure(args['exp']))
main(args)
default="MWE-DGCN",
help="Model to use")
parser.add_argument("-c", "--cuda", type=str, default="none")
parser.add_argument(
"--postfix",
type=str,
default="",
help="a string appended to the file name of the saved model")
parser.add_argument("--rand_seed",
type=int,
default=-1,
help="random seed for torch and numpy")
parser.add_argument("--residual", action="store_true")
parser.add_argument("--ewnorm",
type=str,
default="none",
choices=["none", "both"])
args = parser.parse_args().__dict__
# Get experiment configuration
args["dataset"] = "ogbn-proteins"
args["exp_name"] = "_".join([args["model"], args["dataset"]])
args.update(get_exp_configure(args))
if not (args["cuda"] == "none"):
args["device"] = torch.device("cuda: " + str(args["cuda"]))
else:
args["device"] = torch.device("cpu")
main(args)
