repeat_size=repeat_size,
batch_size=batch_size,
num_parallel_workers=num_parallel_workers,
is_training=is_training)
return cifar_ds
if __name__ == '__main__':
args_opt = parse_args()
context.set_context(mode=context.GRAPH_MODE,
device_target=args_opt.device_target)
# download cifar10 dataset
if not args_opt.dataset_path:
args_opt.dataset_path = download_dataset('cifar10')
# build the network
if args_opt.do_eval and args_opt.load_pretrained == 'hub':
from tinyms import hub
net = hub.load(args_opt.hub_uid, class_num=args_opt.num_classes)
else:
net = vgg16(class_num=args_opt.num_classes)
net.update_parameters_name(prefix='huawei')
model = Model(net)
# define the loss function
net_loss = SoftmaxCrossEntropyWithLogits(sparse=True, reduction="mean")
# define the optimizer
net_opt = Momentum(filter(lambda x: x.requires_grad, net.get_parameters()),
0.01, 0.9)
model.compile(loss_fn=net_loss,
optimizer=net_opt,
mnist_ds,
repeat_size=repeat_size,
batch_size=batch_size,
num_parallel_workers=num_parallel_workers)
return mnist_ds
if __name__ == "__main__":
args_opt = parse_args()
context.set_context(mode=context.GRAPH_MODE,
device_target=args_opt.device_target)
# download mnist dataset
if not args_opt.dataset_path:
args_opt.dataset_path = download_dataset('mnist')
# build the network
net = lenet5()
model = Model(net)
# define the loss function
net_loss = SoftmaxCrossEntropyWithLogits(sparse=True, reduction='mean')
# define the optimizer
net_opt = Momentum(net.trainable_params(), 0.01, 0.9)
model.compile(net_loss, net_opt, metrics={"Accuracy": Accuracy()})
epoch_size = args_opt.epoch_size
batch_size = args_opt.batch_size
mnist_path = args_opt.dataset_path
dataset_sink_mode = not args_opt.device_target == "CPU"
if args_opt.do_eval: # as for evaluation, users could use model.eval
json_dict['categories'].append(cat)
anno_file = os.path.join(anno_dir, 'annotation.json')
with open(anno_file, 'w') as f:
json.dump(json_dict, f)
return anno_file
if __name__ == '__main__':
args_opt = parse_args()
context.set_context(mode=context.GRAPH_MODE,
device_target=args_opt.device_target)
# download voc dataset
if not args_opt.dataset_path:
args_opt.dataset_path = download_dataset('voc')
epoch_size = args_opt.epoch_size
batch_size = args_opt.batch_size
voc_path = args_opt.dataset_path
dataset_sink_mode = not args_opt.device_target == "CPU"
if not args_opt.do_eval: # as for train, users could use model.train
ds_train = create_dataset(voc_path, batch_size=batch_size)
dataset_size = ds_train.get_dataset_size()
# build the SSD300 network
net = ssd300_mobilenetv2(class_num=args_opt.num_classes)
# define the loss function
if args_opt.device_target == "GPU":
net.to_float(ts.float16)
net = net_with_loss(net)
def test_download_dataset_voc():
download_dataset(dataset_name='voc', local_path='/tmp')
assert os.path.exists('/tmp/voc/VOCdevkit/VOC2007')
def test_download_dataset_cifar100():
download_dataset(dataset_name='cifar100', local_path='/tmp')
assert os.path.exists('/tmp/cifar100/cifar-100-bin/train.bin')
assert os.path.exists('/tmp/cifar100/cifar-100-bin/test.bin')
def test_download_dataset_cifar10():
download_dataset(dataset_name='cifar10', local_path='/tmp')
assert os.path.exists('/tmp/cifar10/cifar-10-batches-bin/batches.meta.txt')
def test_download_dataset_mnist():
download_dataset(dataset_name='mnist', local_path='/tmp')
assert os.path.exists('/tmp/mnist/train')
assert os.path.exists('/tmp/mnist/test')
train_ds = kaggle_display_advertising_ds.load_mindreocrd_dataset(
usage='train', batch_size=batch_size)
eval_ds = kaggle_display_advertising_ds.load_mindreocrd_dataset(
usage='test', batch_size=batch_size)
return train_ds, eval_ds
if __name__ == "__main__":
args_opt = parse_args()
context.set_context(mode=context.GRAPH_MODE,
device_target=args_opt.device_target)
# download kaggle display advertising dataset
if not args_opt.dataset_path:
args_opt.dataset_path = download_dataset('kaggle_display_advertising')
else:
args_opt.dataset_path = os.path.join(args_opt.dataset_path,
"kaggle_display_advertising")
epoch_size = args_opt.epoch_size
batch_size = args_opt.batch_size
dataset_path = args_opt.dataset_path
dataset_sink_mode = not args_opt.device_target == "CPU"
checkpoint_dir = args_opt.checkpoint_dir if args_opt.checkpoint_dir is not None else "."
# create train and eval dataset
train_ds, eval_ds = create_dataset(data_path=dataset_path,
batch_size=batch_size)
# build base network
data_size = train_ds.get_dataset_size()
