class TestFeaturePredictorPredict(unittest.TestCase):
def setUp(self):
self.link = FeaturePredictor(DummyFeatureExtractor(
self.in_channels, self.shape_0, self.shape_1),
crop_size=5,
crop=self.crop)
self.x = np.random.uniform(size=(3, self.in_channels, 32,
32)).astype(np.float32)
self.one_output = self.shape_1 is None
def check(self, x):
out = self.link.predict(x)
if self.one_output:
self.assertEqual(out.shape, (self.x.shape[0], ) + self.shape_0)
self.assertIsInstance(out, np.ndarray)
else:
out_0, out_1 = out
self.assertEqual(out_0.shape, (self.x.shape[0], ) + self.shape_0)
self.assertEqual(out_1.shape, (self.x.shape[0], ) + self.shape_1)
self.assertIsInstance(out_0, np.ndarray)
self.assertIsInstance(out_1, np.ndarray)
def test_cpu(self):
self.check(self.x)
@attr.gpu
def test_gpu(self):
self.link.to_gpu()
self.check(self.x)
class TestFeaturePredictorPredict(unittest.TestCase):
def setUp(self):
self.link = FeaturePredictor(
DummyFeatureExtractor(
self.in_channels, self.shape_0, self.shape_1),
crop_size=5, crop=self.crop)
self.x = np.random.uniform(
size=(3, self.in_channels, 32, 32)).astype(np.float32)
self.one_output = self.shape_1 is None
def check(self, x):
out = self.link.predict(x)
if self.one_output:
self.assertEqual(out.shape, (self.x.shape[0],) + self.shape_0)
self.assertIsInstance(out, np.ndarray)
else:
out_0, out_1 = out
self.assertEqual(out_0.shape, (self.x.shape[0],) + self.shape_0)
self.assertEqual(out_1.shape, (self.x.shape[0],) + self.shape_1)
self.assertIsInstance(out_0, np.ndarray)
self.assertIsInstance(out_1, np.ndarray)
def test_cpu(self):
self.check(self.x)
@attr.gpu
def test_gpu(self):
self.link.to_gpu()
self.check(self.x)
def setUp(self):
self.link = FeaturePredictor(DummyFeatureExtractor(
self.in_channels, self.shape_0, self.shape_1),
crop_size=5,
crop=self.crop)
self.x = np.random.uniform(size=(3, self.in_channels, 32,
32)).astype(np.float32)
self.one_output = self.shape_1 is None
def main():
parser = argparse.ArgumentParser(
description='Learning convnet from ILSVRC2012 dataset')
parser.add_argument('val', help='Path to root of the validation dataset')
parser.add_argument(
'--model', choices=('vgg16', 'resnet50', 'resnet101', 'resnet152'))
parser.add_argument('--pretrained_model', default='imagenet')
parser.add_argument('--gpu', type=int, default=-1)
parser.add_argument('--batchsize', type=int, default=32)
parser.add_argument('--crop', choices=('center', '10'), default='center')
parser.add_argument('--resnet_mode', default='he')
args = parser.parse_args()
dataset = DirectoryParsingLabelDataset(args.val)
label_names = directory_parsing_label_names(args.val)
n_class = len(label_names)
iterator = iterators.MultiprocessIterator(
dataset, args.batchsize, repeat=False, shuffle=False,
n_processes=6, shared_mem=300000000)
if args.model == 'vgg16':
extractor = VGG16(n_class, args.pretrained_model)
elif args.model == 'resnet50':
extractor = ResNet50(
n_class, args.pretrained_model, mode=args.resnet_mode)
elif args.model == 'resnet101':
extractor = ResNet101(
n_class, args.pretrained_model, mode=args.resnet_mode)
elif args.model == 'resnet152':
extractor = ResNet152(
n_class, args.pretrained_model, mode=args.resnet_mode)
model = FeaturePredictor(
extractor, crop_size=224, scale_size=256, crop=args.crop)
if args.gpu >= 0:
chainer.cuda.get_device(args.gpu).use()
model.to_gpu()
print('Model has been prepared. Evaluation starts.')
in_values, out_values, rest_values = apply_to_iterator(
model.predict, iterator, hook=ProgressHook(len(dataset)))
del in_values
pred_probs, = out_values
gt_labels, = rest_values
accuracy = F.accuracy(
np.array(list(pred_probs)), np.array(list(gt_labels))).data
print()
print('Top 1 Error {}'.format(1. - accuracy))
def setUp(self):
self.link = FeaturePredictor(
DummyFeatureExtractor(self.in_channels, (1,), None),
crop_size=self.crop_size, scale_size=self.scale_size,
crop=self.crop, mean=self.mean)
if isinstance(self.crop_size, int):
hw = (self.crop_size, self.crop_size)
else:
hw = self.crop_size
if self.crop == 'center':
self.expected_shape = (1, self.in_channels) + hw
elif self.crop == '10':
self.expected_shape = (10, self.in_channels) + hw
class TestFeaturePredictor(unittest.TestCase):
def setUp(self):
self.link = FeaturePredictor(
DummyFeatureExtractor(self.in_channels, (1,), None),
crop_size=self.crop_size, scale_size=self.scale_size,
crop=self.crop, mean=self.mean)
if isinstance(self.crop_size, int):
hw = (self.crop_size, self.crop_size)
else:
hw = self.crop_size
if self.crop == 'center':
self.expected_shape = (1, self.in_channels) + hw
elif self.crop == '10':
self.expected_shape = (10, self.in_channels) + hw
def test_prepare(self):
out = self.link._prepare(
np.random.uniform(size=(self.in_channels, 286, 286)))
self.assertEqual(out.shape, self.expected_shape)
def test_mean(self):
if self.mean is None:
np.testing.assert_equal(self.link.mean, self.link.extractor.mean)
else:
np.testing.assert_equal(self.link.mean, self.mean)
def setUp(self):
self.link = FeaturePredictor(
DummyFeatureExtractor(
self.in_channels, self.shape_0, self.shape_1),
crop_size=5, crop=self.crop)
self.x = np.random.uniform(
size=(3, self.in_channels, 32, 32)).astype(np.float32)
self.one_output = self.shape_1 is None
def setup(dataset, model, pretrained_model, batchsize, val, crop, resnet_arch):
dataset_name = dataset
if dataset_name == 'imagenet':
dataset = DirectoryParsingLabelDataset(val)
label_names = directory_parsing_label_names(val)
def eval_(out_values, rest_values):
pred_probs, = out_values
gt_labels, = rest_values
accuracy = F.accuracy(np.array(list(pred_probs)),
np.array(list(gt_labels))).data
print()
print('Top 1 Error {}'.format(1. - accuracy))
cls, pretrained_models, default_batchsize = models[model][:3]
if pretrained_model is None:
pretrained_model = pretrained_models.get(dataset_name, dataset_name)
if crop is None:
crop = models[model][3]
kwargs = {
'n_class': len(label_names),
'pretrained_model': pretrained_model,
}
if model in ['resnet50', 'resnet101', 'resnet152']:
if resnet_arch is None:
resnet_arch = models[model][4]
kwargs.update({'arch': resnet_arch})
extractor = cls(**kwargs)
model = FeaturePredictor(extractor,
crop_size=224,
scale_size=256,
crop=crop)
if batchsize is None:
batchsize = default_batchsize
return dataset, eval_, model, batchsize
