def main():
args = get_args()
helper.print_script_args_and_info(args)
datasets = dataset_helper.get_dataset_names_with_concept_map(
limit_datasets=args.limit_dataset)
graph_files = []
for dataset in datasets:
label_lookup_files = glob('{}/{}.*.label-lookup.npy'.format(
args.lookup_path, dataset))
for label_lookup_file in label_lookup_files:
if not os.path.exists(label_lookup_file):
print('No lookup file for dataset found: {}'.format(dataset))
continue
graph_files += [
(cache_file, label_lookup_file)
for cache_file in dataset_helper.get_all_cached_graph_datasets(
dataset_name=dataset)
]
print('# Num tasks: {}'.format(len(graph_files)))
Parallel(n_jobs=args.n_jobs)(
delayed(process_dataset)(cache_file, label_lookup_file, args)
for cache_file, label_lookup_file in graph_files)
LOGGER.info('Finished')
def main():
args = get_args()
start_time = time()
helper.print_script_args_and_info(args)
Parallel(n_jobs=args.n_jobs)(
delayed(process_dataset)(dataset_name, args)
for dataset_name in dataset_helper.get_dataset_names_with_concept_map(
limit_datasets=args.limit_dataset))
print('Finished (time={})'.format(
time_utils.seconds_to_human_readable(time() - start_time)))
def main():
args = get_args()
helper.print_script_args_and_info(args)
os.makedirs(args.embeddings_result_folder, exist_ok=True)
LOGGER.info('Loading pre-trained embedding')
LOGGER.info('Starting to process datasets')
Parallel(n_jobs=args.n_jobs)(
delayed(process_dataset)(dataset_name, args)
for dataset_name in dataset_helper.get_dataset_names_with_concept_map(
limit_datasets=args.limit_dataset))
LOGGER.info('Finished')
def main():
args = get_args()
helper.print_script_args_and_info(args)
limited_datasets = args.limit_dataset
os.makedirs(args.embedding_save_path, exist_ok=True)
datasets = dataset_helper.get_dataset_names_with_concept_map(
limit_datasets=limited_datasets)
Parallel(n_jobs=args.n_jobs)(delayed(process_dataset)(dataset_name, args)
for dataset_name in datasets)
LOGGER.info('Finished')
def main():
args = get_args()
helper.print_script_args_and_info(args)
for graph_folder in glob(args.graphs_folder + '/*'):
if graph_folder.rsplit('/', 1)[1].startswith('_'): continue
if not os.path.isdir(graph_folder): continue
print('Processing: {}'.format(graph_folder))
graph_dataset_name = graph_folder.split('/')[-1]
try:
X, Y = dataset_helper.get_gml_graph_dataset(
dataset_name=graph_dataset_name,
graphs_folder=args.graphs_folder,
use_cached=args.use_cached,
suffix=args.suffix
)
except Exception as e:
traceback.print_exc()
print('Error:', e)
def main():
args = get_args()
helper.print_script_args_and_info(args)
if args.experiment_config:
experiment_config = experiment_helper.get_experiment_config(
args.experiment_config)
else:
experiment_config = {}
create_results_dir(args)
classification_options: ClassificationOptions = ClassificationOptions.from_argparse_options(
args)
tasks: typing.List[ExperimentTask] = experiments.get_all_tasks()
start_tasks(args, tasks, classification_options, experiment_config)
def main():
args = get_args()
helper.print_script_args_and_info(args)
Parallel(n_jobs=args.n_jobs)(
delayed(process_graph_cache_file)(graph_cache_file, args)
for graph_cache_file in dataset_helper.get_all_cached_graph_datasets())