def _check(self):
if self.filename is None:
extension = DetectionVisReport(self.iterator, self.link)
else:
extension = DetectionVisReport(self.iterator,
self.link,
filename=self.filename)
if not _available:
return
for iter_ in range(3):
self.trainer.updater.iteration = iter_
extension(self.trainer)
for idx in six.moves.range(len(self.dataset)):
out_file = os.path.join(self.trainer.out,
self.filename_func(iter_, idx))
self.assertTrue(os.path.exists(out_file))
def setUp(self):
self.trainer = mock.MagicMock()
self.out_dir = tempfile.mkdtemp()
self.trainer.out = self.out_dir
self.trainer.updater.iteration = 0
model = ConstantReturnModel(np.random.uniform(size=(1, ) + self.shape))
dataset = DummyDataset(shapes=[(3, 10, 10), self.shape],
dtypes=[np.float32, np.float32])
self.extension = DetectionVisReport(self.indices,
dataset,
model,
filename_base='detection')
class TestDetectionVisReport(unittest.TestCase):
def setUp(self):
self.trainer = mock.MagicMock()
self.trainer.out = tempfile.mkdtemp()
self.link = _RandomDetectionStubLink()
self.dataset = TupleDataset(np.random.uniform(size=(10, 3, 32, 48)),
np.random.uniform(size=(10, 5, 4)),
np.random.randint(0, 19, size=(10, 5)))
self.iterator = SerialIterator(self.dataset,
10,
repeat=False,
shuffle=False)
def test_available(self):
self.extension = DetectionVisReport(self.dataset, self.link)
self.assertEqual(self.extension.available(), _available)
def _check(self):
if self.filename is None:
extension = DetectionVisReport(self.iterator, self.link)
else:
extension = DetectionVisReport(self.iterator,
self.link,
filename=self.filename)
if not _available:
return
for iter_ in range(3):
self.trainer.updater.iteration = iter_
extension(self.trainer)
for idx in six.moves.range(len(self.dataset)):
out_file = os.path.join(self.trainer.out,
self.filename_func(iter_, idx))
self.assertTrue(os.path.exists(out_file))
def test_cpu(self):
self._check()
@attr.gpu
def test_gpu(self):
self.link.to_gpu()
self._check()
class TestDetectionVisReport(unittest.TestCase):
def setUp(self):
self.trainer = mock.MagicMock()
self.trainer.out = tempfile.mkdtemp()
self.link = _RandomDetectionStubLink()
self.dataset = TupleDataset(
np.random.uniform(size=(10, 3, 32, 48)),
np.random.uniform(size=(10, 5, 4)),
np.random.randint(0, 19, size=(10, 5)))
self.iterator = SerialIterator(
self.dataset, 10, repeat=False, shuffle=False)
def test_available(self):
self.extension = DetectionVisReport(self.dataset, self.link)
self.assertEqual(self.extension.available(), optional_modules)
def _check(self):
if self.filename is None:
extension = DetectionVisReport(self.iterator, self.link)
else:
extension = DetectionVisReport(
self.iterator, self.link, filename=self.filename)
if not optional_modules:
return
for iter_ in range(3):
self.trainer.updater.iteration = iter_
extension(self.trainer)
for idx in six.moves.range(len(self.dataset)):
out_file = os.path.join(
self.trainer.out, self.filename_func(iter_, idx))
self.assertTrue(os.path.exists(out_file))
def test_cpu(self):
self._check()
@attr.gpu
def test_gpu(self):
self.link.to_gpu()
self._check()
def test_available(self):
self.extension = DetectionVisReport(self.dataset, self.link)
self.assertEqual(self.extension.available(), _available)
from faster_rcnn import FasterRCNN
if __name__ == '__main__':
test_data = VOCDetectionDataset(mode='train',
use_cache=True,
year='2007',
bgr=True)
wrappers = [
lambda d: SubtractWrapper(d,
value=np.array([103.939, 116.779, 123.68])),
lambda d: ResizeWrapper(d,
preprocess_idx=0,
output_shape=output_shape_soft_min_hard_max(
600, 1200),
hook=bbox_resize_hook(1)),
]
for wrapper in wrappers:
test_data = wrapper(test_data)
model = FasterRCNN()
chainer.serializers.load_npz('VGG16_faster_rcnn_final.model', model)
trainer = mock.MagicMock()
trainer.out = 'result'
trainer.updater.iteration = 0
extension = DetectionVisReport([3, 4, 5, 6, 7, 8],
test_data,
model,
predict_func=model.predict_bboxes)
extension(trainer)
def test_available(self):
self.extension = DetectionVisReport(self.dataset, self.link)
self.assertEqual(self.extension.available(), optional_modules)
def main(gpu=-1, epoch=100, batch_size=1, lr=5e-4, out='result'):
train_data = VOCDetectionDataset(mode='train', use_cache=True, year='2007')
test_data = VOCDetectionDataset(mode='val', use_cache=True, year='2007')
def transform(in_data):
img, bbox = in_data
img -= np.array([103.939, 116.779, 123.68])[:, None, None]
# Resize bounding box to a shape
# with the smaller edge at least at length 600
input_shape = img.shape[1:]
output_shape = _shape_soft_min_hard_max(input_shape, 600, 1200)
img = transforms.resize(img, output_shape)
bbox = transforms.bbox_resize(bbox, input_shape, output_shape)
# horizontally flip
img, flips = transforms.random_flip(img,
horizontal_flip=True,
return_flip=True)
h_flip = flips['h']
bbox = transforms.bbox_flip(bbox, output_shape, h_flip)
return img, bbox
transforms.extend(train_data, transform)
transforms.extend(test_data, transform)
model = FasterRCNN(gpu=gpu)
if gpu != -1:
model.to_gpu(gpu)
chainer.cuda.get_device(gpu).use()
# optimizer = chainer.optimizers.MomentumSGD(lr=lr)
optimizer = chainer.optimizers.Adam(alpha=0.001,
beta1=0.9,
beta2=0.999,
eps=1e-8)
optimizer.setup(model)
# optimizer.add_hook(chainer.optimizer.WeightDecay(rate=0.0005))
train_iter = chainer.iterators.SerialIterator(test_data, batch_size=1)
updater = ParallelUpdater(train_iter, optimizer, devices={'main': gpu})
# updater = chainer.training.updater.StandardUpdater(train_iter, optimizer)
trainer = training.Trainer(updater, (epoch, 'epoch'), out=out)
log_interval = 20, 'iteration'
val_interval = 3000, 'iteration'
trainer.extend(extensions.LogReport(trigger=log_interval))
trainer.extend(extensions.PrintReport([
'iteration', 'main/time', 'main/rpn_loss_cls', 'main/rpn_loss_bbox',
'main/loss_cls', 'main/loss_bbox'
]),
trigger=log_interval)
trainer.extend(extensions.ProgressBar(update_interval=10))
# visualize training
trainer.extend(extensions.PlotReport(['main/rpn_loss_cls'],
file_name='rpn_loss_cls.png'),
trigger=log_interval)
trainer.extend(extensions.PlotReport(['main/rpn_loss_bbox'],
file_name='rpn_loss_bbox.png'),
trigger=log_interval)
trainer.extend(extensions.PlotReport(['main/loss_cls'],
file_name='loss_cls.png'),
trigger=log_interval)
trainer.extend(extensions.PlotReport(['main/loss_bbox'],
file_name='loss_bbox.png'),
trigger=log_interval)
trainer.extend(
DetectionVisReport(
range(10), # visualize outputs for the first 10 data of test_data
train_data,
model,
filename_base='detection_train',
predict_func=model.predict_bboxes),
trigger=val_interval,
invoke_before_training=True)
trainer.extend(
DetectionVisReport(
range(10), # visualize outputs for the first 10 data of test_data
test_data,
model,
forward_func=model.predict_bboxes),
trigger=val_interval,
invoke_before_training=True)
trainer.extend(extensions.dump_graph('main/loss'))
trainer.run()