def test_transforms_presets_rcnn():
im_fname = gcv.utils.download('https://github.com/dmlc/web-data/blob/master/' +
'gluoncv/detection/biking.jpg?raw=true', path='biking.jpg')
x, orig_img = rcnn.load_test(im_fname, short=600, max_size=1000)
x1, orig_img1 = rcnn.transform_test(mx.image.imread(im_fname), short=600, max_size=1000)
np.testing.assert_allclose(x.asnumpy(), x1.asnumpy())
np.testing.assert_allclose(orig_img, orig_img1)
if not osp.isdir(osp.expanduser('~/.mxnet/datasets/voc')):
return
train_dataset = gcv.data.VOCDetection(splits=((2007, 'trainval'), (2012, 'trainval')))
val_dataset = gcv.data.VOCDetection(splits=[(2007, 'test')])
width, height = (512, 512)
net = gcv.model_zoo.get_model('faster_rcnn_resnet50_v1b_voc', pretrained=False, pretrained_base=False)
net.initialize()
num_workers = 0
short, max_size = 600, 1000
batch_size = 4
train_bfn = batchify.Tuple(*[batchify.Append() for _ in range(5)])
train_loader = mx.gluon.data.DataLoader(
train_dataset.transform(rcnn.FasterRCNNDefaultTrainTransform(short, max_size, net)),
batch_size, True, batchify_fn=train_bfn, last_batch='rollover', num_workers=num_workers)
val_bfn = batchify.Tuple(*[batchify.Append() for _ in range(3)])
val_loader = mx.gluon.data.DataLoader(
val_dataset.transform(rcnn.FasterRCNNDefaultValTransform(short, max_size)),
batch_size, False, batchify_fn=val_bfn, last_batch='keep', num_workers=num_workers)
train_loader2 = gluon.data.DataLoader(
train_dataset.transform(rcnn.FasterRCNNDefaultTrainTransform(short, max_size)),
batch_size, True, batchify_fn=batchify.Tuple(*[batchify.Append() for _ in range(2)]),
last_batch='rollover', num_workers=num_workers)
for loader in [train_loader, val_loader, train_loader2]:
for i, batch in enumerate(loader):
if i > 1:
break
pass
def test_transforms_presets_rcnn():
im_fname = gcv.utils.download('https://github.com/dmlc/web-data/blob/master/' +
'gluoncv/detection/biking.jpg?raw=true', path='biking.jpg')
x, orig_img = rcnn.load_test(im_fname, short=600, max_size=1000)
x1, orig_img1 = rcnn.transform_test(mx.image.imread(im_fname), short=600, max_size=1000)
np.testing.assert_allclose(x.asnumpy(), x1.asnumpy())
np.testing.assert_allclose(orig_img, orig_img1)
if not osp.isdir(osp.expanduser('~/.mxnet/datasets/voc')):
return
train_dataset = VOCDetectionTiny()
val_dataset = VOCDetectionTiny(splits=[('tiny_motorbike', 'test')])
width, height = (512, 512)
net = gcv.model_zoo.get_model('faster_rcnn_resnet50_v1b_voc', pretrained=False, pretrained_base=False)
net.initialize()
num_workers = 0
short, max_size = 600, 1000
batch_size = 4
train_bfn = batchify.Tuple(*[batchify.Append() for _ in range(5)])
train_loader = mx.gluon.data.DataLoader(
train_dataset.transform(rcnn.FasterRCNNDefaultTrainTransform(short, max_size, net, flip_p=0.5)),
batch_size, True, batchify_fn=train_bfn, last_batch='rollover', num_workers=num_workers)
val_bfn = batchify.Tuple(*[batchify.Append() for _ in range(3)])
val_loader = mx.gluon.data.DataLoader(
val_dataset.transform(rcnn.FasterRCNNDefaultValTransform(short, max_size)),
batch_size, False, batchify_fn=val_bfn, last_batch='keep', num_workers=num_workers)
train_loader2 = gluon.data.DataLoader(
train_dataset.transform(rcnn.FasterRCNNDefaultTrainTransform(short, max_size)),
batch_size, True, batchify_fn=batchify.Tuple(*[batchify.Append() for _ in range(2)]),
last_batch='rollover', num_workers=num_workers)
for loader in [train_loader, val_loader, train_loader2]:
for i, batch in enumerate(loader):
if i > 1:
break
pass
