def test_fail_on_mismatching_dimensionality(self):
from lasagne.layers.pool import MaxPool2DLayer
with pytest.raises(ValueError) as exc:
MaxPool2DLayer((10, 20, 30), 3, 2)
assert "Expected 4 input dimensions" in exc.value.args[0]
with pytest.raises(ValueError) as exc:
MaxPool2DLayer((10, 20, 30, 40, 50), 3, 2)
assert "Expected 4 input dimensions" in exc.value.args[0]
def layer(self, input_layer, ds, st=None):
from lasagne.layers.pool import MaxPool2DLayer
return MaxPool2DLayer(
input_layer,
ds=ds,
st=st,
ignore_border=False,
)
def layer_ignoreborder(self, input_layer, ds, st=None, pad=(0, 0)):
from lasagne.layers.pool import MaxPool2DLayer
return MaxPool2DLayer(
input_layer,
ds=ds,
st=st,
pad=pad,
ignore_border=True,
)
def layer(self, input_layer, pool_size, stride=None,
pad=(0, 0), ignore_border=False):
from lasagne.layers.pool import MaxPool2DLayer
return MaxPool2DLayer(
input_layer,
pool_size=pool_size,
stride=stride,
pad=pad,
ignore_border=ignore_border,
)
def build_model(inp):
net = {}
net['input'] = InputLayer((None, 3, 224, 224), input_var=inp)
net['conv1'] = Conv2DLayer(net['input'],
96,
7,
stride=2,
pad=(1, 1),
W=lasagne.init.Uniform())
net['norm1'] = LocalResponseNormalization2DLayer(net['conv1'],
alpha=0.00005)
net['pool1'] = MaxPool2DLayer(net['norm1'], 3, stride=2)
net['conv2'] = Conv2DLayer(net['pool1'],
256,
5,
stride=2,
pad=(0, 0),
W=lasagne.init.Uniform(),
b=lasagne.init.Constant(1.))
net['norm2'] = LocalResponseNormalization2DLayer(net['conv2'],
alpha=0.00005)
net['pool2'] = MaxPool2DLayer(net['norm2'], 3, stride=2)
net['conv3'] = Conv2DLayer(net['pool2'],
384,
3,
stride=1,
pad=(1, 1),
W=lasagne.init.Uniform())
net['conv4'] = Conv2DLayer(net['conv3'],
384,
3,
stride=1,
pad=(1, 1),
W=lasagne.init.Uniform(),
b=lasagne.init.Constant(1.))
net['conv5'] = Conv2DLayer(net['conv4'],
256,
3,
stride=1,
pad=(1, 1),
W=lasagne.init.Uniform(),
b=lasagne.init.Constant(1.))
net['p5_spm6'] = FlattenLayer(MaxPool2DLayer(net['conv5'], 3, stride=2))
net['p5_spm3'] = FlattenLayer(MaxPool2DLayer(net['conv5'], 5, stride=4))
net['p5_spm2'] = FlattenLayer(MaxPool2DLayer(net['conv5'], 7, stride=7))
net['p5_spm1'] = MaxPool2DLayer(net['conv5'], 13, stride=13)
net['p5_spm1_f'] = FlattenLayer(net['p5_spm1'])
net['conc_spm'] = ConcatLayer(
[net['p5_spm6'], net['p5_spm3'], net['p5_spm2'], net['p5_spm1_f']])
net['fc6'] = DenseLayer(net['conc_spm'],
num_units=4096,
W=lasagne.init.Uniform(),
b=lasagne.init.Constant(1.))
net['fc6_drop'] = DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = DenseLayer(net['fc6_drop'],
num_units=4096,
W=lasagne.init.Uniform(),
b=lasagne.init.Constant(1.))
net['fc7_drop'] = DropoutLayer(net['fc7'], p=0.5)
net['fc8'] = DenseLayer(net['fc6_drop'],
num_units=1000,
W=lasagne.init.Uniform(),
b=lasagne.init.Constant(1.))
net['out'] = NonlinearityLayer(net['fc8'], softmax)
return net
def layer_ignoreborder(self, input_layer):
from lasagne.layers.pool import MaxPool2DLayer
return MaxPool2DLayer(input_layer, ds=(2, 2), ignore_border=True)
def layer(self, input_layer):
from lasagne.layers.pool import MaxPool2DLayer
return MaxPool2DLayer(input_layer, ds=(2, 2))
def build_model(inp):
net = {}
net['input'] = InputLayer((None, 3, 224, 224), input_var=input_var)
net['conv1_1'] = batch_norm(
Conv2DLayer(net['input'], 64, 3, pad=1, flip_filters=False))
net['conv1_2'] = batch_norm(
Conv2DLayer(net['conv1_1'], 64, 3, pad=1, flip_filters=False))
net['pool1'] = MaxPool2DLayer(net['conv1_2'], 2)
net['conv2_1'] = batch_norm(
Conv2DLayer(net['pool1'], 128, 3, pad=1, flip_filters=False))
net['conv2_2'] = batch_norm(
Conv2DLayer(net['conv2_1'], 128, 3, pad=1, flip_filters=False))
net['pool2'] = MaxPool2DLayer(net['conv2_2'], 2)
net['conv3_1'] = batch_norm(
Conv2DLayer(net['pool2'], 256, 3, pad=1, flip_filters=False))
net['conv3_2'] = batch_norm(
Conv2DLayer(net['conv3_1'], 256, 3, pad=1, flip_filters=False))
net['conv3_3'] = batch_norm(
Conv2DLayer(net['conv3_2'], 256, 3, pad=1, flip_filters=False))
net['pool3'] = MaxPool2DLayer(net['conv3_3'], 2)
net['conv4_1'] = batch_norm(
Conv2DLayer(net['pool3'], 512, 3, pad=1, flip_filters=False))
net['conv4_2'] = batch_norm(
Conv2DLayer(net['conv4_1'], 512, 3, pad=1, flip_filters=False))
net['conv4_3'] = batch_norm(
Conv2DLayer(net['conv4_2'], 512, 3, pad=1, flip_filters=False))
net['pool4'] = MaxPool2DLayer(net['conv4_3'], 2)
net['conv5_1'] = batch_norm(
Conv2DLayer(net['pool4'], 512, 3, pad=1, flip_filters=False))
net['conv5_2'] = batch_norm(
Conv2DLayer(net['conv5_1'], 512, 3, pad=1, flip_filters=False))
net['conv5_3'] = batch_norm(
Conv2DLayer(net['conv5_2'], 512, 3, pad=1, flip_filters=False))
net['pool5'] = MaxPool2DLayer(net['conv5_3'], 2)
net['fc6'] = batch_norm(DenseLayer(net['pool5'], num_units=4096))
net['fc6_dropout'] = DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = batch_norm(DenseLayer(net['fc6_dropout'], num_units=4096))
net['fc7_dropout'] = DropoutLayer(net['fc7'], p=0.5)
net['reshape'] = ReshapeLayer(net['fc7_dropout'], (-1, 4096, 1, 1))
net['deconv6'] = batch_norm(Deconv2DLayer(net['reshape'], 512, 7))
net['unpool5'] = Upscale2DLayer(net['deconv6'], 2)
net['deconv5_3'] = batch_norm(
Deconv2DLayer(net['unpool5'], 512, 3, crop=1, flip_filters=False))
net['deconv5_2'] = batch_norm(
Deconv2DLayer(net['deconv5_3'], 512, 3, crop=1, flip_filters=False))
net['deconv5_1'] = batch_norm(
Deconv2DLayer(net['deconv5_2'], 512, 3, crop=1, flip_filters=False))
net['unpool4'] = Upscale2DLayer(net['deconv5_1'], 2)
net['deconv4_3'] = batch_norm(
Deconv2DLayer(net['unpool4'], 512, 3, crop=1, flip_filters=False))
net['deconv4_2'] = batch_norm(
Deconv2DLayer(net['deconv4_3'], 512, 3, crop=1, flip_filters=False))
net['deconv4_1'] = batch_norm(
Deconv2DLayer(net['deconv4_2'], 512, 3, crop=1, flip_filters=False))
net['unpool3'] = Upscale2DLayer(net['deconv4_1'], 2)
net['deconv3_3'] = batch_norm(
Deconv2DLayer(net['unpool3'], 256, 3, crop=1, flip_filters=False))
net['deconv3_2'] = batch_norm(
Deconv2DLayer(net['deconv3_3'], 256, 3, crop=1, flip_filters=False))
net['deconv3_1'] = batch_norm(
Deconv2DLayer(net['deconv3_2'], 256, 3, crop=1, flip_filters=False))
net['unpool2'] = Upscale2DLayer(net['deconv3_1'], 2)
net['deconv2_2'] = batch_norm(
Deconv2DLayer(net['unpool2'], 128, 3, crop=1, flip_filters=False))
net['deconv2_1'] = batch_norm(
Deconv2DLayer(net['deconv2_2'], 128, 3, crop=1, flip_filters=False))
net['unpool1'] = Upscale2DLayer(net['deconv2_1'], 2)
net['deconv1_1'] = batch_norm(
Deconv2DLayer(net['unpool1'], 64, 3, crop=1, flip_filters=False))
net['deconv1_2'] = batch_norm(
Deconv2DLayer(net['deconv1_1'], 64, 3, crop=1, flip_filters=False))
net['out'] = NonlinearityLayer(DenseLayer(net['deconv1_2'], 21), softmax)
return net