def __init__(self, n_layer, pretrained_model=None, initialW=None):
n_block = self._blocks[n_layer]
_, path = utils.prepare_pretrained_model({}, pretrained_model,
self._models[n_layer])
super(DilatedResNet, self).__init__()
with self.init_scope():
self.conv1_1 = Conv2DBNActiv(None,
64,
3,
2,
1,
1,
initialW=initialW)
self.conv1_2 = Conv2DBNActiv(64, 64, 3, 1, 1, 1, initialW=initialW)
self.conv1_3 = Conv2DBNActiv(64,
128,
3,
1,
1,
1,
initialW=initialW)
self.pool1 = lambda x: F.max_pooling_2d(
x, ksize=3, stride=2, pad=1)
self.res2 = ResBlock(n_block[0],
128,
64,
256,
1,
1,
initialW=initialW,
stride_first=False)
self.res3 = ResBlock(n_block[1],
256,
128,
512,
2,
1,
initialW=initialW,
stride_first=False)
self.res4 = ResBlock(n_block[2],
512,
256,
1024,
1,
2,
initialW=initialW,
stride_first=False)
self.res5 = ResBlock(n_block[3],
1024,
512,
2048,
1,
4,
initialW=initialW,
stride_first=False)
if path:
chainer.serializers.load_npz(path, self, ignore_names=None)
def __init__(self,
out_size,
transform_rois_to_grayscale=False,
train_imagenet=False):
super(SheepLocalizer, self).__init__()
initialW = initializers.HeNormal(scale=1., fan_option='fan_out')
kwargs = {"initialW": initialW}
keys_to_remove = ['fc6', 'prob'] if not train_imagenet else ['prob']
with self.init_scope():
self.feature_extractor = MyResNet50Layers(
keys_to_remove=keys_to_remove, pretrained_model='auto')
if not train_imagenet:
self.param_predictor = L.Linear(2048, 6)
self.res6 = ResBlock(2, None, 1024, 2048, 2, **kwargs)
self.res7 = ResBlock(2, None, 1024, 2048, 2, **kwargs)
transform_bias = self.param_predictor.b.data
transform_bias[[0, 4]] = 0.8
transform_bias[[2, 5]] = 0
self.param_predictor.W.data[...] = 0
self.visual_backprop_anchors = []
self.out_size = out_size
self.transform_rois_to_grayscale = transform_rois_to_grayscale
self.train_imagenet = train_imagenet
def __init__(self, n_layer,
n_class=None,
pretrained_model=None,
mean=None, initialW=None, fc_kwargs={}):
blocks = self._blocks[n_layer]
self.mean = mean
if initialW is None:
initialW = initializers.HeNormal(scale=1., fan_option='fan_out')
if 'initialW' not in fc_kwargs:
fc_kwargs['initialW'] = initializers.Normal(scale=0.01)
kwargs = {
'groups': 32, 'initialW': initialW, 'stride_first': False,
'add_seblock': True}
super(SEResNeXt, self).__init__()
with self.init_scope():
self.conv1 = Conv2DBNActiv(None, 64, 3, 1, 1, nobias=True,
initialW=initialW)
self.res2 = ResBlock(blocks[0], None, 128, 256, 2, **kwargs)
self.res3 = ResBlock(blocks[1], None, 256, 512, 1, **kwargs)
self.res4 = ResBlock(blocks[2], None, 512, 1024, 2, **kwargs)
self.res5 = ResBlock(blocks[3], None, 1024, 2048, 1, **kwargs)
self.pool5 = lambda x: F.average(x, axis=(2, 3))
self.fc6 = L.Linear(None, n_class, **fc_kwargs)
def __init__(self,
n_layer,
n_class=None,
pretrained_model=None,
mean=None,
initialW=None,
fc_kwargs={}):
blocks = self._blocks[n_layer]
param, path = utils.prepare_pretrained_model(
{
'n_class': n_class,
'mean': mean
}, pretrained_model, self._models[n_layer], {
'n_class': 1000,
'mean': _imagenet_mean
})
self.mean = param['mean']
if initialW is None:
initialW = initializers.HeNormal(scale=1., fan_option='fan_out')
if 'initialW' not in fc_kwargs:
fc_kwargs['initialW'] = initializers.Normal(scale=0.01)
if pretrained_model:
# As a sampling process is time-consuming,
# we employ a zero initializer for faster computation.
initialW = initializers.constant.Zero()
fc_kwargs['initialW'] = initializers.constant.Zero()
kwargs = {
'groups': 32,
'initialW': initialW,
'stride_first': False,
'add_seblock': True
}
super(SEResNeXt, self).__init__()
with self.init_scope():
self.conv1 = Conv2DBNActiv(None,
64,
7,
2,
3,
nobias=True,
initialW=initialW)
self.pool1 = lambda x: F.max_pooling_2d(x, ksize=3, stride=2)
self.res2 = ResBlock(blocks[0], None, 128, 256, 1, **kwargs)
self.res3 = ResBlock(blocks[1], None, 256, 512, 2, **kwargs)
self.res4 = ResBlock(blocks[2], None, 512, 1024, 2, **kwargs)
self.res5 = ResBlock(blocks[3], None, 1024, 2048, 2, **kwargs)
self.pool5 = lambda x: F.average(x, axis=(2, 3))
self.fc6 = L.Linear(None, param['n_class'], **fc_kwargs)
self.prob = F.softmax
if path:
chainer.serializers.load_npz(path, self)
def __init__(self, n_layer, initialW, bn_kwargs=None):
n_block = self._blocks[n_layer]
super(DilatedResNet, self).__init__()
with self.init_scope():
self.conv1_1 = Conv2DBNActiv(None,
64,
3,
2,
1,
1,
initialW=initialW,
bn_kwargs=bn_kwargs)
self.conv1_2 = Conv2DBNActiv(64,
64,
3,
1,
1,
1,
initialW=initialW,
bn_kwargs=bn_kwargs)
self.conv1_3 = Conv2DBNActiv(64,
128,
3,
1,
1,
1,
initialW=initialW,
bn_kwargs=bn_kwargs)
self.pool1 = lambda x: F.max_pooling_2d(
x, ksize=3, stride=2, pad=1)
self.res2 = ResBlock(n_block[0],
128,
64,
256,
1,
1,
initialW=initialW,
bn_kwargs=bn_kwargs,
stride_first=False)
self.res3 = ResBlock(n_block[1],
256,
128,
512,
2,
1,
initialW=initialW,
bn_kwargs=bn_kwargs,
stride_first=False)
self.res4 = ResBlock(n_block[2],
512,
256,
1024,
1,
2,
initialW=initialW,
bn_kwargs=bn_kwargs,
stride_first=False)
self.res5 = ResBlock(n_block[3],
1024,
512,
2048,
1,
4,
initialW=initialW,
bn_kwargs=bn_kwargs,
stride_first=False)