def parse_args():
parser = argparse.ArgumentParser(
description="Train classification models on ImageNet",
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
models.add_model_args(parser)
fit.add_fit_args(parser)
data.add_data_args(parser)
dali.add_dali_args(parser)
data.add_data_aug_args(parser)
return parser.parse_args()
warm_up_done = True
except StopIteration as e:
print("get_data exception due to end of data - resetting iterator")
it.reset()
db = it.next()
finally:
yield get_arrays(db)
parser = argparse.ArgumentParser(description="train_resnet50",
formatter_class=argparse.ArgumentDefaultsHelpFormatter)
data.add_data_args(parser)
data.add_data_aug_args(parser)
parser.add_argument_group('gpu_config', 'gpu config')
parser.add_argument('--num-gpus', type=int, default=8, help='Number of GPUs to use during training')
parser.add_argument('--batch-per-gpu', type=int, default=64, help='Batch size per GPU')
# Use a large augmentation level (needed by ResNet-50)
data.set_data_aug_level(parser, 3)
parser.set_defaults(
# network
network = 'resnet',
num_layers = 50,
# data
num_classes = 1000,
num_examples = 1281167,
image_shape = '3,224,224',