def __init__(self, weights=None, augmentation=False):
super(GoogLeNet, self).__init__(weights, augmentation)
def build_inception_module(name, input_layer, nfilters):
# nfilters: (pool_proj, 1x1, 3x3_reduce, 3x3, 5x5_reduce, 5x5)
net = {}
net['pool'] = PoolLayerDNN(input_layer, pool_size=3, stride=1, pad=1)
net['pool_proj'] = ConvLayer(net['pool'], nfilters[0], 1, flip_filters=False)
net['1x1'] = ConvLayer(input_layer, nfilters[1], 1, flip_filters=False)
net['3x3_reduce'] = ConvLayer(input_layer, nfilters[2], 1, flip_filters=False)
net['3x3'] = ConvLayer(net['3x3_reduce'], nfilters[3], 3, pad=1, flip_filters=False)
net['5x5_reduce'] = ConvLayer(input_layer, nfilters[4], 1, flip_filters=False)
net['5x5'] = ConvLayer(net['5x5_reduce'], nfilters[5], 5, pad=2, flip_filters=False)
net['output'] = lasagne.layers.ConcatLayer([
net['1x1'],
net['3x3'],
net['5x5'],
net['pool_proj']])
return {'{}/{}'.format(name, k): v for k, v in net.items()}
net = {}
net['input'] = lasagne.layers.InputLayer((None, 3, 224, 224))
net['conv1/7x7_s2'] = ConvLayer(net['input'], 64, 7, stride=2, pad=3, flip_filters=False)
net['pool1/3x3_s2'] = PoolLayer(net['conv1/7x7_s2'], pool_size=3, stride=2, ignore_border=False)
net['pool1/norm1'] = lasagne.layers.LocalResponseNormalization2DLayer(net['pool1/3x3_s2'], alpha=0.00002, k=1)
net['conv2/3x3_reduce'] = ConvLayer(net['pool1/norm1'], 64, 1, flip_filters=False)
net['conv2/3x3'] = ConvLayer(net['conv2/3x3_reduce'], 192, 3, pad=1, flip_filters=False)
net['conv2/norm2'] = lasagne.layers.LocalResponseNormalization2DLayer(net['conv2/3x3'], alpha=0.00002, k=1)
net['pool2/3x3_s2'] = PoolLayer(net['conv2/norm2'], pool_size=3, stride=2, ignore_border=False)
net.update(build_inception_module('inception_3a', net['pool2/3x3_s2'], [32, 64, 96, 128, 16, 32]))
net.update(build_inception_module('inception_3b', net['inception_3a/output'], [64, 128, 128, 192, 32, 96]))
net['pool3/3x3_s2'] = PoolLayer(net['inception_3b/output'], pool_size=3, stride=2, ignore_border=False)
net.update(build_inception_module('inception_4a', net['pool3/3x3_s2'], [64, 192, 96, 208, 16, 48]))
net.update(build_inception_module('inception_4b', net['inception_4a/output'], [64, 160, 112, 224, 24, 64]))
net.update(build_inception_module('inception_4c', net['inception_4b/output'], [64, 128, 128, 256, 24, 64]))
net.update(build_inception_module('inception_4d', net['inception_4c/output'], [64, 112, 144, 288, 32, 64]))
net.update(build_inception_module('inception_4e', net['inception_4d/output'], [128, 256, 160, 320, 32, 128]))
net['pool4/3x3_s2'] = PoolLayer(net['inception_4e/output'], pool_size=3, stride=2, ignore_border=False)
net.update(build_inception_module('inception_5a', net['pool4/3x3_s2'], [128, 256, 160, 320, 32, 128]))
net.update(build_inception_module('inception_5b', net['inception_5a/output'], [128, 384, 192, 384, 48, 128]))
net['pool5/7x7_s1'] = lasagne.layers.GlobalPoolLayer(net['inception_5b/output'])
net['dropout5'] = lasagne.layers.DropoutLayer(net['pool5/7x7_s1'], p=0.4)
self.net = net
self.out_layer = net['dropout5']
if self.weights is not None:
init_weights = self._get_weights_from_file(self.weights, 'param values')
init_weights = init_weights[:-2] # since we have chopped off the last two layers of the network (loss3/classifier and prob), we won't need those
lasagne.layers.set_all_param_values(self.out_layer, init_weights)
def __init__(self, weights=None, augmentation=False):
super(VGG16, self).__init__(weights, augmentation)
net = {}
net['input'] = lasagne.layers.InputLayer((None, 3, 224, 224))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1, flip_filters=False)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1, flip_filters=False)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1, flip_filters=False)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1, flip_filters=False)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1, flip_filters=False)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1, flip_filters=False)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1, flip_filters=False)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1, flip_filters=False)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1, flip_filters=False)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1, flip_filters=False)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1, flip_filters=False)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1, flip_filters=False)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1, flip_filters=False)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = lasagne.layers.DenseLayer(net['pool5'], num_units=4096)
net['fc6_dropout'] = lasagne.layers.DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = lasagne.layers.DenseLayer(net['fc6_dropout'], num_units=4096)
net['fc7_dropout'] = lasagne.layers.DropoutLayer(net['fc7'], p=0.5)
self.net = net
self.out_layer = net['fc7_dropout']
if self.weights is not None:
init_weights = self._get_weights_from_file(self.weights, 'param values')
init_weights = init_weights[:-2] # since we have chopped off the last two layers of the network, we won't need those
lasagne.layers.set_all_param_values(self.out_layer, init_weights)
def conv_pool_cnn_c():
net = {}
net['input'] = InputLayer((None, 3, 32, 32))
net['drop_in'] = DropoutLayer(net['input'], p=0.2)
net['conv1_1'] = ConvLayer(net['drop_in'],
num_filters=96,
filter_size=3,
pad=1,
flip_filters=False)
net['conv1_2'] = ConvLayer(net['conv1_1'],
num_filters=96,
filter_size=3,
pad=1,
flip_filters=False)
net['conv1_3'] = ConvLayer(net['conv1_2'],
num_filters=96,
filter_size=3,
pad=1,
flip_filters=False)
net['conv2_1'] = PoolLayer(net['conv1_3'], pool_size=3, stride=2)
net['drop2_1'] = DropoutLayer(net['conv2_1'], p=0.5)
net['conv3_1'] = ConvLayer(net['drop2_1'],
num_filters=192,
filter_size=3,
pad=1,
flip_filters=False)
net['conv3_2'] = ConvLayer(net['conv3_1'],
num_filters=192,
filter_size=3,
pad=1,
flip_filters=False)
net['conv3_3'] = ConvLayer(net['conv3_2'],
num_filters=192,
filter_size=3,
pad=1,
flip_filters=False)
net['conv4_1'] = PoolLayer(net['conv3_3'], pool_size=3, stride=2)
net['drop4_1'] = DropoutLayer(net['conv4_1'], p=0.5)
net['conv5_1'] = ConvLayer(net['drop4_1'],
num_filters=192,
filter_size=3,
pad=1,
flip_filters=False)
net['conv6_1'] = ConvLayer(net['conv5_1'],
num_filters=192,
filter_size=1,
flip_filters=False)
net['conv7_1'] = ConvLayer(net['conv6_1'],
num_filters=10,
filter_size=1,
flip_filters=False)
net['global_avg'] = GlobalPoolLayer(net['conv7_1'])
net['output'] = NonlinearityLayer(net['global_avg'], softmax)
return net
def initialize_network(width):
# initialize network - VGG19 style
net = {}
net['input'] = InputLayer((1, 3, width, width))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_4'], 2, mode='average_exc_pad')
return net
def define_vgg19_base(input_var=None, nonlinearity=rectify):
net = {}
net['input'] = InputLayer((None, 3, None, None), input_var=input_var)
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
layersstr = ['conv1_1','conv1_2',
'conv2_1','conv2_2',
'conv3_1','conv3_2','conv3_3','conv3_4',
'conv4_1','conv4_2','conv4_3','conv4_4',
'conv5_1','conv5_2','conv5_3','conv5_4']
return net, layersstr
def build_model(self,numclasses):
#NOTE: perhaps we should save the network architecture along with the weights, since they could be different
logger.info("Building network")
net = {}
net['input'] = InputLayer((None, 3, 224, 224))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['fc7'] = DenseLayer(net['fc6'], num_units=4096)
net['fc8'] = DenseLayer(net['fc7'], num_units=numclasses, nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
logger.info("Network built")
return net
def build_small_model():
net = {}
net['input_layer'] = InputLayer((None,3,128,128))
net['conv1_1'] = ConvLayer(
net['input_layer'], 64, 3, pad=1)#, flip_filters=False)
net['conv1_2'] = ConvLayer(
net['conv1_1'], 64, 3, pad=1)#, flip_filters=False)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(
net['pool1'], 128, 3, pad=1)#, flip_filters=False)
net['conv2_2'] = ConvLayer(
net['conv2_1'], 128, 3, pad=1 )#, flip_filters=False)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(
net['pool2'], 256, 3, pad=1 )#, flip_filters=False)
net['conv3_2'] = ConvLayer(
net['conv3_1'], 256, 3, pad=1)#, flip_filters=False)
net['pool3'] = PoolLayer(net['conv3_2'], 2)
net['conv4_1'] = ConvLayer(
net['pool3'], 512, 3, pad=1)#, flip_filters=False)
net['pool4'] = PoolLayer(net['conv4_1'], 2)
net['fc6'] = DenseLayer(net['pool4'], num_units=200)
net['fc6_dropout'] = DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = DenseLayer(net['fc6_dropout'], num_units=100)
net['fc7_dropout'] = DropoutLayer(net['fc7'], p=0.5)
net['fc8'] = DenseLayer(
net['fc7_dropout'], num_units=2, nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
return net
def build_model(input_var,dro=0.5):
net = {}
net['input'] = InputLayer((None, 3, 299, 299),input_var=input_var)
print(net['input'])
net['conv1/7x7_s2'] = ConvLayer(
net['input'], 64, 7, stride=2, pad=3, flip_filters=False)
print(net['conv1/7x7_s2'])
net['pool1/3x3_s2'] = PoolLayer(
net['conv1/7x7_s2'], pool_size=3, stride=2, ignore_border=False)
net['pool1/norm1'] = LRNLayer(net['pool1/3x3_s2'], alpha=0.00002, k=1)
net['conv2/3x3_reduce'] = ConvLayer(
net['pool1/norm1'], 64, 1, flip_filters=False)
net['conv2/3x3'] = ConvLayer(
net['conv2/3x3_reduce'], 192, 3, pad=1, flip_filters=False)
net['conv2/norm2'] = LRNLayer(net['conv2/3x3'], alpha=0.00002, k=1)
net['pool2/3x3_s2'] = PoolLayer(
net['conv2/norm2'], pool_size=3, stride=2, ignore_border=False)
net.update(build_inception_module('inception_3a',
net['pool2/3x3_s2'],
[32, 64, 96, 128, 16, 32]))
net.update(build_inception_module('inception_3b',
net['inception_3a/output'],
[64, 128, 128, 192, 32, 96]))
net['pool3/3x3_s2'] = PoolLayer(
net['inception_3b/output'], pool_size=3, stride=2, ignore_border=False)
net.update(build_inception_module('inception_4a',
net['pool3/3x3_s2'],
[64, 192, 96, 208, 16, 48],dro))
net.update(build_inception_module('inception_4b',
net['inception_4a/output'],
[64, 160, 112, 224, 24, 64],dro))
net.update(build_inception_module('inception_4c',
net['inception_4b/output'],
[64, 128, 128, 256, 24, 64],dro))
net.update(build_inception_module('inception_4d',
net['inception_4c/output'],
[64, 112, 144, 288, 32, 64],dro))
net.update(build_inception_module('inception_4e',
net['inception_4d/output'],
[128, 256, 160, 320, 32, 128],dro))
net['pool4/3x3_s2'] = PoolLayer(
net['inception_4e/output'], pool_size=3, stride=2, ignore_border=False)
net.update(build_inception_module('inception_5a',
net['pool4/3x3_s2'],
[128, 256, 160, 320, 32, 128]))
net.update(build_inception_module('inception_5b',
net['inception_5a/output'],
[128, 384, 192, 384, 48, 128]))
net['pool5/7x7_s1'] = GlobalPoolLayer(net['inception_5b/output'])
net['pool5/7x7_s1_dropout'] = DropoutLayer(net['pool5/7x7_s1'], p=dro)
net['loss3/classifier'] = DenseLayer(net['pool5/7x7_s1_dropout'],
num_units=1,
nonlinearity=linear)
net['prob'] = NonlinearityLayer(net['loss3/classifier'],
nonlinearity=sigmoid)
return net
def load_vgg19(pkl_filename="vgg19_normalized.pkl"):
net = {}
net['input'] = InputLayer((1, 3, None, None))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_4'], 2, mode='average_exc_pad')
if bytes == str:
# Python 2
values = pickle.load(open(pkl_filename, 'rb'))['param values']
else:
#Python 3
values = pickle.load(open(pkl_filename, 'rb'),
encoding='latin1')['param values']
set_all_param_values(net['pool5'], values[:32])
return net
def build_model():
net = {}
net['input'] = InputLayer((None, 3, 224, 224))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['drop6'] = DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = DenseLayer(net['drop6'], num_units=4096)
net['drop7'] = DropoutLayer(net['fc7'], p=0.5)
net['fc8'] = DenseLayer(net['drop7'], num_units=1000, nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
return net
def build_model():
net = {}
net['input'] = InputLayer((1, 3, None, None))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_4'], 2, mode='average_exc_pad')
return net
def CNN_model():
net = {}
net['input'] = InputLayer((None, 3, 64, 64))
net['conv1/3x3_s1'] = ConvLayer(net['input'], 16, 3, stride=2, pad='same', W=lasagne.init.HeNormal(gain='relu')) # 16*64*64
# net['pool1/3x3_s2'] = PoolLayer(net['conv1/3x3_s1'], pool_size=3, stride=2, ignore_border=False) # 16*32*32
net['pool1/norm1'] = LRNLayer(net['conv1/3x3_s2'], alpha=0.00002, k=1)
net['conv2/3x3_s1'] = ConvLayer(net['pool1/norm1'], 32, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu')) # 32*32*32
net['pool2/3x3_s2'] = PoolLayer(net['conv2/3x3_s1'], pool_size=3, stride=2, ignore_border=False) # 32*16*16
net['pool2/norm1'] = LRNLayer(net['pool2/3x3_s2'], alpha=0.00002, k=1)
net['conv3/3x3_s1'] = ConvLayer(net['pool2/norm1'], 64, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu')) # 64*16*16
net['pool3/3x3_s2'] = PoolLayer(net['conv3/3x3_s1'], pool_size=3, stride=2, ignore_border=False) # 64*8*8
net['pool3/norm1'] = LRNLayer(net['pool3/3x3_s2'], alpha=0.00002, k=1)
net['conv4/3x3_s1'] = ConvLayer(net['pool3/norm1'], 128, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu')) # 128*8*8
net['pool4/3x3_s2'] = PoolLayer(net['conv4/3x3_s1'], pool_size=3, stride=2, ignore_border=False) # 128*4*4
net['pool4/norm1'] = LRNLayer(net['pool4/3x3_s2'], alpha=0.00002, k=1)
net['conv5/3x3_s1'] = ConvLayer(net['pool4/norm1'], 128, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu'))
net['pool5/norm1'] = LRNLayer(net['conv5/3x3_s1'], alpha=0.00002, k=1)
net['conv6/3x3_s1'] = ConvLayer(net['pool5/norm1'], 128, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu'))
net['pool6/norm1'] = LRNLayer(net['conv6/3x3_s1'], alpha=0.00002, k=1)
net['conv7/3x3_s1'] = ConvLayer(net['pool6/norm1'], 128, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu'))
net['pool7/norm1'] = LRNLayer(net['conv7/3x3_s1'], alpha=0.00002, k=1)
net['conv8/3x3_s1'] = ConvLayer(net['pool7/norm1'], 256, 3, stride=1, pad='same', W=lasagne.init.HeNormal(gain='relu')) # 256*4*4
net['pool8/4x4_s1'] = GlobalPoolLayer(net['conv8/3x3_s1']); # 256
return net
def build_vgg(shape, input_var):
net = {}
w, h = shape
net['input'] = InputLayer((None, 3, w, h), input_var=input_var)
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_4'], 2, mode='average_exc_pad')
return net
def build_model(timesteps, pX, pY):
net = OrderedDict()
net['input'] = InputLayer(shape=(None, timesteps, pX, pY))
net['conv1_1'] = batch_norm(
ConvLayer(net['input'],
num_filters=32,
filter_size=(3, 3),
nonlinearity=ReLU,
pad='same'))
net['conv1_2'] = batch_norm(
ConvLayer(net['conv1_1'],
num_filters=32,
filter_size=(3, 3),
nonlinearity=ReLU,
pad='same'))
net['pool1'] = PoolLayer(net['conv1_2'], pool_size=(2, 2))
net['conv2_1'] = batch_norm(
ConvLayer(net['pool1'],
num_filters=64,
filter_size=(3, 3),
nonlinearity=ReLU,
pad='same'))
net['conv2_2'] = batch_norm(
ConvLayer(net['conv2_1'],
num_filters=64,
filter_size=(3, 3),
nonlinearity=ReLU,
pad='same'))
net['pool2'] = PoolLayer(net['conv2_2'], pool_size=(2, 2))
net['conv3_1'] = batch_norm(
ConvLayer(net['pool2'],
num_filters=128,
filter_size=(3, 3),
nonlinearity=ReLU,
pad='same'))
net['conv3_2'] = batch_norm(
ConvLayer(net['conv3_1'],
num_filters=128,
filter_size=(3, 3),
nonlinearity=ReLU,
pad='same'))
net['pool3'] = PoolLayer(net['conv3_2'], pool_size=(2, 2))
net['fcLayer1'] = batch_norm(
DenseLayer(dropout(net['pool3'], p=0.5),
num_units=512,
nonlinearity=ReLU))
net['output'] = DenseLayer(dropout(net['fcLayer1'], p=0.5),
num_units=2,
nonlinearity=softmax)
return net
def setup_perceptual(self, input):
"""Use lasagne to create a network of convolution layers using pre-trained VGG19 weights.
"""
offset = np.array([103.939, 116.779, 123.680], dtype=np.float32).reshape((1,3,1,1))
self.network['percept'] = lasagne.layers.NonlinearityLayer(input, lambda x: ((x+0.5)*255.0) - offset)
self.network['mse'] = self.network['percept']
self.network['conv1_1'] = ConvLayer(self.network['percept'], 64, 3, pad=1)
self.network['conv1_2'] = ConvLayer(self.network['conv1_1'], 64, 3, pad=1)
self.network['pool1'] = PoolLayer(self.network['conv1_2'], 2, mode='max')
self.network['conv2_1'] = ConvLayer(self.network['pool1'], 128, 3, pad=1)
self.network['conv2_2'] = ConvLayer(self.network['conv2_1'], 128, 3, pad=1)
self.network['pool2'] = PoolLayer(self.network['conv2_2'], 2, mode='max')
self.network['conv3_1'] = ConvLayer(self.network['pool2'], 256, 3, pad=1)
self.network['conv3_2'] = ConvLayer(self.network['conv3_1'], 256, 3, pad=1)
self.network['conv3_3'] = ConvLayer(self.network['conv3_2'], 256, 3, pad=1)
self.network['conv3_4'] = ConvLayer(self.network['conv3_3'], 256, 3, pad=1)
self.network['pool3'] = PoolLayer(self.network['conv3_4'], 2, mode='max')
self.network['conv4_1'] = ConvLayer(self.network['pool3'], 512, 3, pad=1)
self.network['conv4_2'] = ConvLayer(self.network['conv4_1'], 512, 3, pad=1)
self.network['conv4_3'] = ConvLayer(self.network['conv4_2'], 512, 3, pad=1)
self.network['conv4_4'] = ConvLayer(self.network['conv4_3'], 512, 3, pad=1)
self.network['pool4'] = PoolLayer(self.network['conv4_4'], 2, mode='max')
self.network['conv5_1'] = ConvLayer(self.network['pool4'], 512, 3, pad=1)
self.network['conv5_2'] = ConvLayer(self.network['conv5_1'], 512, 3, pad=1)
self.network['conv5_3'] = ConvLayer(self.network['conv5_2'], 512, 3, pad=1)
self.network['conv5_4'] = ConvLayer(self.network['conv5_3'], 512, 3, pad=1)
def get_net(self):
net = {}
net['input'] = InputLayer((None, 3, 224, 224))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['fc6_dropout'] = DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = DenseLayer(net['fc6_dropout'], num_units=4096)
net['fc7_dropout'] = DropoutLayer(net['fc7'], p=0.5)
net['fc8'] = DenseLayer(net['fc7_dropout'],
num_units=1000,
nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
output_layer = net['prob']
return net, net['prob'] # the whole net and the output layer
def build_model():
print("building model")
net = {}
net['input'] = InputLayer((1, 3, IMAGE_W, IMAGE_W))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_4'], 2, mode='average_exc_pad')
print("model built.")
print()
print("Loading weights.")
values = pickle.load(open('vgg19_normalized.pkl'))['param values']
print()
print("Loaded. Setting the weights.")
lasagne.layers.set_all_param_values(net['pool5'], values)
print()
print("Set.")
return net
def build_vgg(self):
''' Builds the VGG-16, 16-layer model as given in the paper:
"Very Deep Convolutional Networks for Large-Scale Image Recognition"
and assigns optimal weights from the ImageNet classification challenge dataset
Original source: https://gist.github.com/ksimonyan/211839e770f7b538e2d8'''
net = {}
net['input'] = InputLayer((None, 3, 224, 224))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['fc7'] = DenseLayer(net['fc6'], num_units=4096)
net['fc8'] = DenseLayer(net['fc7'], num_units=1000, nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
lasagne.layers.set_all_param_values(net['prob'],
self.vgg16_params['param values'])
print "Built VGG-16 model"
return net
def build_vgg_model():
net = {}
net['input'] = InputLayer((1, 3, IMAGE_W, IMAGE_W))
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1, flip_filters=False)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1, flip_filters=False)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1, flip_filters=False)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1, flip_filters=False)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1, flip_filters=False)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1, flip_filters=False)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1, flip_filters=False)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1, flip_filters=False)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1, flip_filters=False)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1, flip_filters=False)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1, flip_filters=False)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1, flip_filters=False)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1, flip_filters=False)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1, flip_filters=False)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1, flip_filters=False)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1, flip_filters=False)
net['pool5'] = PoolLayer(net['conv5_4'], 2, mode='average_exc_pad')
return net
def makeNeuralNet(input_var=None):
net = {}
net['input'] = InputLayer(shape=(None, 3, 224, 224), input_var=input_var)
net['bnorm'] = BatchNormLayer(net['input'])
net['conv1'] = ConvLayer(net['bnorm'],
num_filters=96,
filter_size=5,
stride=2) #96*112*112
net['norm1'] = NormLayer(
net['conv1'], alpha=0.0001) # caffe has alpha = alpha * pool_size
net['pool1'] = PoolLayer(net['norm1'],
pool_size=3,
stride=3,
ignore_border=False) #96*37...approx
net['conv2'] = ConvLayer(net['pool1'],
num_filters=256,
filter_size=5,
pad=1)
net['pool2'] = PoolLayer(net['conv2'],
pool_size=2,
stride=2,
ignore_border=False)
net['fc6'] = DenseLayer(net['pool2'], num_units=1024)
net['drop6'] = DropoutLayer(net['fc6'], p=0.2)
net['_fc7'] = DenseLayer(net['drop6'], num_units=256)
net['_drop7'] = DropoutLayer(net['_fc7'], p=0.2)
net['_fc8out'] = DenseLayer(net['_drop7'],
num_units=1,
nonlinearity=lasagne.nonlinearities.sigmoid)
output_layer_driver = net['_fc8out']
return output_layer_driver, net
def build_model_vgg16_3channels(input_var=None):
net = {}
net['input'] = InputLayer((None, 3, 224, 224), input_var=input_var)
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['fc7'] = DenseLayer(net['fc6'], num_units=4096)
net['fc8'] = DenseLayer(net['fc7'],
num_units=num_categories,
nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
return net['prob']
def build_model(input_var):
net = {}
net['input'] = InputLayer((None, 3, 224, 224), input_var=input_var)
net['conv1_1'] = ConvLayer(net['input'], 64, 3, pad=1, flip_filters=False)
net['conv1_2'] = ConvLayer(net['conv1_1'],
64,
3,
pad=1,
flip_filters=False)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1, flip_filters=False)
net['conv2_2'] = ConvLayer(net['conv2_1'],
128,
3,
pad=1,
flip_filters=False)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1, flip_filters=False)
net['conv3_2'] = ConvLayer(net['conv3_1'],
256,
3,
pad=1,
flip_filters=False)
net['conv3_3'] = ConvLayer(net['conv3_2'],
256,
3,
pad=1,
flip_filters=False)
net['pool3'] = PoolLayer(net['conv3_3'], 2)
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1, flip_filters=False)
net['conv4_2'] = ConvLayer(net['conv4_1'],
512,
3,
pad=1,
flip_filters=False)
net['conv4_3'] = ConvLayer(net['conv4_2'],
512,
3,
pad=1,
flip_filters=False)
net['pool4'] = PoolLayer(net['conv4_3'], 2)
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1, flip_filters=False)
net['conv5_2'] = ConvLayer(net['conv5_1'],
512,
3,
pad=1,
flip_filters=False)
net['conv5_3'] = ConvLayer(net['conv5_2'],
512,
3,
pad=1,
flip_filters=False)
net['pool5'] = PoolLayer(net['conv5_3'], 2)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['fc7'] = DenseLayer(net['fc6'], num_units=4096)
net['fc8'] = DenseLayer(net['fc7'], num_units=1000, nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc8'], softmax)
return net
def build_model_LeNet(input_var=None):
net = {}
net['input'] = InputLayer(shape=(None, 1, 224, 224), input_var=input_var)
net['conv1/7x7_s2'] = ConvLayer(net['input'], 64, 7, stride=2, pad=3)
net['pool1/3x3_s2'] = PoolLayer(net['conv1/7x7_s2'],
pool_size=3,
stride=2,
ignore_border=False)
net['pool1/norm1'] = LRNLayer(net['pool1/3x3_s2'], alpha=0.00002, k=1)
net['conv2/3x3_reduce'] = ConvLayer(net['pool1/norm1'], 64, 1)
net['conv2/3x3'] = ConvLayer(net['conv2/3x3_reduce'], 192, 3, pad=1)
net['conv2/norm2'] = LRNLayer(net['conv2/3x3'], alpha=0.00002, k=1)
net['pool2/3x3_s2'] = PoolLayer(net['conv2/norm2'], pool_size=3, stride=2)
net.update(
build_inception_module('inception_3a', net['pool2/3x3_s2'],
[32, 64, 96, 128, 16, 32]))
net.update(
build_inception_module('inception_3b', net['inception_3a/output'],
[64, 128, 128, 192, 32, 96]))
net['pool3/3x3_s2'] = PoolLayer(net['inception_3b/output'],
pool_size=3,
stride=2)
net.update(
build_inception_module('inception_4a', net['pool3/3x3_s2'],
[64, 192, 96, 208, 16, 48]))
net.update(
build_inception_module('inception_4b', net['inception_4a/output'],
[64, 160, 112, 224, 24, 64]))
net.update(
build_inception_module('inception_4c', net['inception_4b/output'],
[64, 128, 128, 256, 24, 64]))
net.update(
build_inception_module('inception_4d', net['inception_4c/output'],
[64, 112, 144, 288, 32, 64]))
net.update(
build_inception_module('inception_4e', net['inception_4d/output'],
[128, 256, 160, 320, 32, 128]))
net['pool4/3x3_s2'] = PoolLayer(net['inception_4e/output'],
pool_size=3,
stride=2)
net.update(
build_inception_module('inception_5a', net['pool4/3x3_s2'],
[128, 256, 160, 320, 32, 128]))
net.update(
build_inception_module('inception_5b', net['inception_5a/output'],
[128, 384, 192, 384, 48, 128]))
net['pool5/7x7_s1'] = GlobalPoolLayer(net['inception_5b/output'])
net['loss3/classifier'] = DenseLayer(net['pool5/7x7_s1'],
num_units=num_categories,
nonlinearity=linear)
net['prob'] = NonlinearityLayer(net['loss3/classifier'],
nonlinearity=softmax)
#print(net)
return net['prob']
def build_resnet():
net = {}
net['input'] = InputLayer((None, 3, 224, 224))
sub_net, parent_layer_name = build_simple_block(
net['input'], ['conv1', 'bn_conv1', 'conv1_relu'],
64, 7, 2, 3, use_bias=True)
net.update(sub_net)
net['pool1'] = PoolLayer(net[parent_layer_name], pool_size=3, stride=2,
pad=0, mode='max', ignore_border=False)
block_size = list('abc')
parent_layer_name = 'pool1'
for c in block_size:
if c == 'a':
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1, 1, True, 4, ix='2%s' % c)
else:
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/4, 1, False, 4, ix='2%s' % c)
net.update(sub_net)
block_size = list('abcd')
for c in block_size:
if c == 'a':
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/2, 1.0/2, True, 4, ix='3%s' % c)
else:
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/4, 1, False, 4, ix='3%s' % c)
net.update(sub_net)
block_size = list('abcdef')
for c in block_size:
if c == 'a':
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/2, 1.0/2, True, 4, ix='4%s' % c)
else:
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/4, 1, False, 4, ix='4%s' % c)
net.update(sub_net)
block_size = list('abc')
for c in block_size:
if c == 'a':
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/2, 1.0/2, True, 4, ix='5%s' % c)
else:
sub_net, parent_layer_name = build_residual_block(net[parent_layer_name],
1.0/4, 1, False, 4, ix='5%s' % c)
net.update(sub_net)
net['pool5'] = PoolLayer(net[parent_layer_name], pool_size=7, stride=1, pad=0,
mode='average_exc_pad', ignore_border=False)
net['fc1000'] = DenseLayer(net['pool5'], num_units=1000, nonlinearity=None)
net['prob'] = NonlinearityLayer(net['fc1000'], nonlinearity=softmax)
return net
def setup_model(self, input=None):
"""Use lasagne to create a network of convolution layers, first using VGG19 as the framework
and then adding augmentations for Semantic Style Transfer.
"""
net, self.channels = {}, {}
# Primary network for the main image. These are convolution only, and stop at layer 4_2 (rest unused).
net['img'] = input or InputLayer((None, 3, None, None))
net['conv1_1'] = ConvLayer(net['img'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['main'] = net['conv5_4']
# Auxiliary network for the semantic layers, and the nearest neighbors calculations.
net['map'] = InputLayer((1, 1, None, None))
for j, i in itertools.product(range(5), range(4)):
if j < 2 and i > 1: continue
suffix = '%i_%i' % (j + 1, i + 1)
if i == 0:
net['map%i' % (j + 1)] = PoolLayer(net['map'],
2**j,
mode='average_exc_pad')
self.channels[suffix] = net['conv' + suffix].num_filters
if args.semantic_weight > 0.0:
net['sem' + suffix] = ConcatLayer(
[net['conv' + suffix], net['map%i' % (j + 1)]])
else:
net['sem' + suffix] = net['conv' + suffix]
net['dup' + suffix] = InputLayer(net['sem' + suffix].output_shape)
net['nn' + suffix] = ConvLayer(net['dup' + suffix],
1,
3,
b=None,
pad=0,
flip_filters=False)
self.network = net
def lasagne_layers_method(self):
'''
INPUT: None
OUTPUT: Dict
Creates dictionary of vgg_cnn_s model Lasagne layer objects. Here the
original output layer (softmax, 1000 classes) has been removed and
the output layer returns a vector of shape (1,4096).
'''
# Create dictionary of VGG_CNN_S model layers
self.lasagne_layers = {}
self.lasagne_layers['input'] = InputLayer((None, 3, 224, 224))
self.lasagne_layers['conv1'] = ConvLayer(self.lasagne_layers['input'],
num_filters=96,
filter_size=7,
stride=2,
flip_filters=False)
self.lasagne_layers['norm1'] = NormLayer(self.lasagne_layers['conv1'],
alpha=0.0001)
self.lasagne_layers['pool1'] = PoolLayer(self.lasagne_layers['norm1'],
pool_size=3,
stride=3,
ignore_border=False)
self.lasagne_layers['conv2'] = ConvLayer(self.lasagne_layers['pool1'],
num_filters=256,
filter_size=5,
flip_filters=False)
self.lasagne_layers['pool2'] = PoolLayer(self.lasagne_layers['conv2'],
pool_size=2,
stride=2,
ignore_border=False)
self.lasagne_layers['conv3'] = ConvLayer(self.lasagne_layers['pool2'],
num_filters=512,
filter_size=3,
pad=1,
flip_filters=False)
self.lasagne_layers['conv4'] = ConvLayer(self.lasagne_layers['conv3'],
num_filters=512,
filter_size=3,
pad=1,
flip_filters=False)
self.lasagne_layers['conv5'] = ConvLayer(self.lasagne_layers['conv4'],
num_filters=512,
filter_size=3,
pad=1,
flip_filters=False)
self.lasagne_layers['pool5'] = PoolLayer(self.lasagne_layers['conv5'],
pool_size=3,
stride=3,
ignore_border=False)
self.lasagne_layers['fc6'] = DenseLayer(self.lasagne_layers['pool5'],
num_units=4096)
self.lasagne_layers['drop6'] = DropoutLayer(self.lasagne_layers['fc6'],
p=0.5)
self.lasagne_layers['fc7'] = DenseLayer(self.lasagne_layers['drop6'],
num_units=4096)
def Deconv_Tiny_VGG(params,input_var=None):
net = {}
net['input'] = InputLayer(shape=(None, 3, 224, 224),input_var=input_var)
net['conv1_1'] = ConvLayer(
net['input'], 32, 3, pad=1, W=params[0], b=params[1], flip_filters=False)
net['conv1_2'] = ConvLayer(
net['conv1_1'], 32, 3, pad=1, W=params[2], b=params[3], flip_filters=False)
net['pool1'] = PoolLayer(net['conv1_2'], 2)
net['conv2_1'] = ConvLayer(
net['pool1'], 32, 3, pad=1, W=params[4], b=params[5], flip_filters=False)
net['conv2_2'] = ConvLayer(
net['conv2_1'], 32, 3, pad=1, W=params[6], b=params[7], flip_filters=False)
net['pool2'] = PoolLayer(net['conv2_2'], 2)
net['conv3_1'] = ConvLayer(
net['pool2'], 32, 3, pad=1, W=params[8], b=params[9], flip_filters=False)
net['conv3_2'] = ConvLayer(
net['conv3_1'], 32, 3, pad=1, W=params[10], b=params[11], flip_filters=False)
net['pool3'] = PoolLayer(net['conv3_2'], 2)
#deconvolution starts here
net['unpool3'] = InverseLayer(net['pool3'],net['pool3'])
net['deconv3_2'] = DeconvLayer(net['unpool3'],num_filters=32,
filter_size=net['conv3_2'].filter_size, stride=net['conv3_2'].stride,
crop=net['conv3_2'].pad,
W=params[10], b=params[9], flip_filters=True)
net['deconv3_1'] = DeconvLayer(net['deconv3_2'],num_filters=32,
filter_size=net['conv3_1'].filter_size, stride=net['conv3_1'].stride,
crop=net['conv3_1'].pad,
W=params[8], b=params[7], flip_filters=True)
net['unpool2'] = InverseLayer(net['deconv3_1'],net['pool2'])
net['deconv2_2'] = DeconvLayer(net['unpool2'],num_filters=32,
filter_size=net['conv2_2'].filter_size, stride=net['conv2_2'].stride,
crop=net['conv2_2'].pad,
W=params[6], b=params[5], flip_filters=True)
net['deconv2_1'] = DeconvLayer(net['deconv2_2'],num_filters=32,
filter_size=net['conv2_1'].filter_size, stride=net['conv2_1'].stride,
crop=net['conv2_1'].pad,
W=params[4], b=params[3], flip_filters=True)
net['unpool1'] = InverseLayer(net['deconv2_1'],net['pool1'])
net['deconv1_2'] = DeconvLayer(net['unpool1'],num_filters=32,
filter_size=net['conv1_2'].filter_size, stride=net['conv1_2'].stride,
crop=net['conv1_2'].pad,
W=params[2], b=params[1], flip_filters=True)
net['deconv1_1'] = DeconvLayer(net['deconv1_2'],num_filters=3,
filter_size=net['conv1_1'].filter_size, stride=net['conv1_1'].stride,
crop=net['conv1_1'].pad,
W=params[0], flip_filters=True)
return net
def network_sm(ip_size, input_var):
net = {}
net['input'] = lasagne.layers.InputLayer(shape=(None, 1, ip_size[0], ip_size[1]), input_var=input_var)
net['conv1'] = ConvLayer(net['input'], 16, 3, pad=0)
net['pool1'] = PoolLayer(net['conv1'], 2)
net['conv2'] = ConvLayer(net['pool1'], 16, 3, pad=0)
net['pool2'] = PoolLayer(net['conv2'], 2)
net['fc1'] = DenseLayer(lasagne.layers.dropout(net['pool2'], p=0.5), num_units=64, nonlinearity=lasagne.nonlinearities.rectify)
net['prob'] = DenseLayer(lasagne.layers.dropout(net['fc1'], p=0.5), num_units=2, nonlinearity=lasagne.nonlinearities.softmax)
return net
def build_model(input_shape):
net = {}
net['input'] = InputLayer(input_shape)
net['conv1'] = ConvLayer(net['input'],
num_filters=96,
filter_size=7,
stride=2,
flip_filters=False)
net['norm1'] = LRNLayer(
net['conv1'], alpha=0.0001) # caffe has alpha = alpha * pool_size
net['pool1'] = PoolLayer(net['norm1'],
pool_size=3,
stride=3,
ignore_border=False)
net['conv2'] = ConvLayer(net['pool1'],
num_filters=256,
filter_size=5,
flip_filters=False)
net['pool2'] = PoolLayer(net['conv2'],
pool_size=2,
stride=2,
ignore_border=False)
net['conv3'] = ConvLayer(net['pool2'],
num_filters=512,
filter_size=3,
pad=1,
flip_filters=False)
net['conv4'] = ConvLayer(net['conv3'],
num_filters=512,
filter_size=3,
pad=1,
flip_filters=False)
net['conv5'] = ConvLayer(net['conv4'],
num_filters=512,
filter_size=3,
pad=1,
flip_filters=False)
net['pool5'] = PoolLayer(net['conv5'],
pool_size=3,
stride=3,
ignore_border=False)
net['fc6'] = DenseLayer(net['pool5'], num_units=4096)
net['drop6'] = DropoutLayer(net['fc6'], p=0.5)
net['fc7'] = DenseLayer(net['drop6'], num_units=4096)
net['drop7'] = DropoutLayer(net['fc7'], p=0.5)
net['fc8'] = DenseLayer(net['drop7'],
num_units=1000,
nonlinearity=lasagne.nonlinearities.softmax)
for layer in net.values():
print str(lasagne.layers.get_output_shape(layer))
return net
def setup_model(self):
"""Use lasagne to create a network of convolution layers, first using VGG19 as the framework
and then adding augmentations for Semantic Style Transfer.
"""
net = {}
# Primary network for the main image. These are convolution only, and stop at layer 4_2 (rest unused).
net['img'] = InputLayer((1, 3, None, None))
net['conv1_1'] = ConvLayer(net['img'], 64, 3, pad=1)
net['conv1_2'] = ConvLayer(net['conv1_1'], 64, 3, pad=1)
net['pool1'] = PoolLayer(net['conv1_2'], 2, mode='average_exc_pad')
net['conv2_1'] = ConvLayer(net['pool1'], 128, 3, pad=1)
net['conv2_2'] = ConvLayer(net['conv2_1'], 128, 3, pad=1)
net['pool2'] = PoolLayer(net['conv2_2'], 2, mode='average_exc_pad')
net['conv3_1'] = ConvLayer(net['pool2'], 256, 3, pad=1)
net['conv3_2'] = ConvLayer(net['conv3_1'], 256, 3, pad=1)
net['conv3_3'] = ConvLayer(net['conv3_2'], 256, 3, pad=1)
net['conv3_4'] = ConvLayer(net['conv3_3'], 256, 3, pad=1)
net['pool3'] = PoolLayer(net['conv3_4'], 2, mode='average_exc_pad')
net['conv4_1'] = ConvLayer(net['pool3'], 512, 3, pad=1)
net['conv4_2'] = ConvLayer(net['conv4_1'], 512, 3, pad=1)
net['conv4_3'] = ConvLayer(net['conv4_2'], 512, 3, pad=1)
net['conv4_4'] = ConvLayer(net['conv4_3'], 512, 3, pad=1)
net['pool4'] = PoolLayer(net['conv4_4'], 2, mode='average_exc_pad')
net['conv5_1'] = ConvLayer(net['pool4'], 512, 3, pad=1)
net['conv5_2'] = ConvLayer(net['conv5_1'], 512, 3, pad=1)
net['conv5_3'] = ConvLayer(net['conv5_2'], 512, 3, pad=1)
net['conv5_4'] = ConvLayer(net['conv5_3'], 512, 3, pad=1)
net['main'] = net['conv5_4']
# Auxiliary network for the semantic layers, and the nearest neighbors calculations.
net['map'] = InputLayer((1, 3, None, None))
for j, i in itertools.product(range(5), range(4)):
if j < 2 and i > 1: continue
suffix = '%i_%i' % (j + 1, i + 1)
net['map' + suffix] = PoolLayer(net['map'],
2**j,
mode='average_exc_pad')
net['nn' + suffix] = ConvLayer(net['conv' + suffix],
1,
3,
b=None,
pad=0)
net['mm' + suffix] = ConvLayer(net['map' + suffix],
1,
3,
b=None,
pad=0)
self.network = net
