def CNNS_model():
if regularization > 0:
W_regularizer = l2(regularization)
b_regularizer = l2(regularization)
else:
W_regularizer = None
b_regularizer = None
# Input Block
inputs = Input(shape=(img_width, img_height, 3), name='input')
if stddev > 0:
inputs = GaussianNoise(stddev=stddev)(inputs)
x = Convolution2D(96, (7, 7), border_mode='same', subsample=2, W_regularizer=l2(regularization),
b_regularizer=l2(regularization), name='conv_1')(inputs)
if LRN2D_norm:
x = LRN2D(alpha=alpha, beta=beta)(x)
x = Activation('relu')(x)
x = MaxPooling2D(pool_size=(3, 3), name='max_pooling_1')(x)
# Conv1 Block
x = Convolution2D(256, (5, 5), border_mode='same', W_regularizer=W_regularizer,
b_regularizer=b_regularizer, name='conv_2')(x)
x = Activation('relu')(x)
x = MaxPooling2D(pool_size=(2, 2), name='max_pooling_2')(x)
# Conv2 Block
x = Convolution2D(512, (3, 3), border_mode='same', W_regularizer=W_regularizer,
b_regularizer=b_regularizer, name='conv_3')(x)
x = Activation('relu')(x)
x = Convolution2D(512, (3, 3), border_mode='same', W_regularizer=W_regularizer,
b_regularizer=b_regularizer, name='conv_4')(x)
x = Activation('relu')(x)
x = Convolution2D(512, (3, 3), border_mode='same', W_regularizer=W_regularizer,
b_regularizer=b_regularizer, name='conv_5')(x)
x = Activation('relu')(x)
x = MaxPooling2D((3, 3), name='max_pooling_3')(x)
x = Flatten()(x)
# Dense Block
x = Dense(4096, activation='relu', name='full6')(x)
if dropout > 0:
x = Dropout(dropout)(x)
x = Dense(4096, activation='relu', name='full7')(x)
if dropout > 0:
x = Dropout(dropout)(x)
# Predictions Block
x = Dense(8, activation='softmax', name='full8')(x)
model = Model(inputs, x, name='cnns_model')
return model
def conv2D_lrn2d(x,filters,kernel_size,strides=(1,1),padding='same',data_format=DATA_FORMAT,dilation_rate=(1,1),activation='relu',use_bias=True,kernel_initializer='glorot_uniform',bias_initializer='zeros',kernel_regularizer=None,bias_regularizer=None,activity_regularizer=None,kernel_constraint=None,bias_constraint=None,lrn2d_norm=LRN2D_NORM,weight_decay=WEIGHT_DECAY):
if weight_decay:
kernel_regularizer=regularizers.l2(weight_decay)
bias_regularizer=regularizers.l2(weight_decay)
else:
kernel_regularizer=None
bias_regularizer=None
x=Conv2D(filters=filters,kernel_size=kernel_size,strides=strides,padding=padding,data_format=data_format,dilation_rate=dilation_rate,activation=activation,use_bias=use_bias,kernel_initializer=kernel_initializer,bias_initializer=bias_initializer,kernel_regularizer=kernel_regularizer,bias_regularizer=bias_regularizer,activity_regularizer=activity_regularizer,kernel_constraint=kernel_constraint,bias_constraint=bias_constraint)(x)
if lrn2d_norm:
x=LRN2D(alpha=ALPHA,beta=BETA)(x)
return x
def conv2D_lrn2d(x,
nb_filter,
nb_row,
nb_col,
border_mode='same',
subsample=(1, 1),
activation='relu',
LRN2D_norm=True,
weight_decay=WEIGHT_DECAY,
dim_ordering=DIM_ORDERING):
'''
Info:
Function taken from the Inceptionv3.py script keras github
Utility function to apply to a tensor a module Convolution + lrn2d
with optional weight decay (L2 weight regularization).
'''
if weight_decay:
W_regularizer = regularizers.l2(weight_decay)
b_regularizer = regularizers.l2(weight_decay)
else:
W_regularizer = None
b_regularizer = None
x = Convolution2D(nb_filter,
nb_row,
nb_col,
subsample=subsample,
activation=activation,
border_mode=border_mode,
W_regularizer=W_regularizer,
b_regularizer=b_regularizer,
bias=False,
dim_ordering=dim_ordering)(x)
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
if LRN2D_norm:
x = LRN2D(alpha=ALPHA, beta=BETA)(x)
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
return x
def create_model():
if DATA_FORMAT == 'channels_first':
INP_SHAPE = (3, 224, 224)
img_input = Input(shape=INP_SHAPE)
CONCAT_AXIS = 1
elif DATA_FORMAT == 'channels_last':
INP_SHAPE = (224, 224, 3)
img_input = Input(shape=INP_SHAPE)
CONCAT_AXIS = 3
else:
raise Exception('Invalid Dim Ordering: ' + str(DIM_ORDERING))
x = conv2D_lrn2d(img_input,
64, (7, 7),
2,
padding='same',
lrn2d_norm=False)
x = MaxPooling2D(pool_size=(3, 3),
strides=2,
padding='same',
data_format=DATA_FORMAT)(x)
x = LRN2D(alpha=ALPHA, beta=BETA)(x)
x = conv2D_lrn2d(x, 64, (1, 1), 1, padding='same', lrn2d_norm=False)
x = conv2D_lrn2d(x, 192, (3, 3), 1, padding='same', lrn2d_norm=True)
x = MaxPooling2D(pool_size=(3, 3),
strides=2,
padding='same',
data_format=DATA_FORMAT)(x)
x = inception_module(x,
params=[(64, ), (96, 128), (16, 32), (32, )],
concat_axis=CONCAT_AXIS) #3a
x = inception_module(x,
params=[(128, ), (128, 192), (32, 96), (64, )],
concat_axis=CONCAT_AXIS) #3b
x = MaxPooling2D(pool_size=(3, 3),
strides=2,
padding='same',
data_format=DATA_FORMAT)(x)
x = inception_module(x,
params=[(192, ), (96, 208), (16, 48), (64, )],
concat_axis=CONCAT_AXIS) #4a
x = inception_module(x,
params=[(160, ), (112, 224), (24, 64), (64, )],
concat_axis=CONCAT_AXIS) #4b
x = inception_module(x,
params=[(128, ), (128, 256), (24, 64), (64, )],
concat_axis=CONCAT_AXIS) #4c
x = inception_module(x,
params=[(112, ), (144, 288), (32, 64), (64, )],
concat_axis=CONCAT_AXIS) #4d
x = inception_module(x,
params=[(256, ), (160, 320), (32, 128), (128, )],
concat_axis=CONCAT_AXIS) #4e
x = MaxPooling2D(pool_size=(3, 3),
strides=2,
padding='same',
data_format=DATA_FORMAT)(x)
x = inception_module(x,
params=[(256, ), (160, 320), (32, 128), (128, )],
concat_axis=CONCAT_AXIS) #5a
x = inception_module(x,
params=[(384, ), (192, 384), (48, 128), (128, )],
concat_axis=CONCAT_AXIS) #5b
x = AveragePooling2D(pool_size=(7, 7),
strides=1,
padding='valid',
data_format=DATA_FORMAT)(x)
x = Flatten()(x)
x = Dropout(DROPOUT)(x)
x = Dense(output_dim=NB_CLASS, activation='linear')(x)
x = Dense(output_dim=NB_CLASS, activation='softmax')(x)
return x, img_input, CONCAT_AXIS, INP_SHAPE, DATA_FORMAT
def create_model():
# Define image input layer
if DIM_ORDERING == 'th':
INP_SHAPE = (3, 224, 224) # 3 - Number of RGB Colours
img_input = Input(shape=INP_SHAPE)
CONCAT_AXIS = 1
elif DIM_ORDERING == 'tf':
INP_SHAPE = (224, 224, 3) # 3 - Number of RGB Colours
img_input = Input(shape=INP_SHAPE)
CONCAT_AXIS = 3
else:
raise Exception('Invalid dim ordering: ' + str(DIM_ORDERING))
# Channel 1 - Convolution Net Input Layer
x = conv2D_lrn2d(img_input,
3,
11,
11,
subsample=(1, 1),
border_mode='same')
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
# Channel 1 - Convolution Net Layer 1
x = conv2D_lrn2d(x, 96, 55, 55, subsample=(1, 1), border_mode='same')
x = MaxPooling2D(strides=(2, 2),
pool_size=(2, 2),
dim_ordering=DIM_ORDERING)(x)
x = LRN2D(alpha=ALPHA, beta=BETA)(x)
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
# Channel 1 - Convolution Net Layer 2
x = conv2D_lrn2d(x, 192, 27, 27, subsample=(1, 1), border_mode='same')
x = MaxPooling2D(strides=(2, 2),
pool_size=(2, 2),
dim_ordering=DIM_ORDERING)(x)
x = LRN2D(alpha=ALPHA, beta=BETA)(x)
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
# Channel 1 - Convolution Net Layer 3
x = conv2D_lrn2d(x, 288, 13, 13, subsample=(1, 1), border_mode='same')
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
# Channel 1 - Convolution Net Layer 4
x = conv2D_lrn2d(x, 288, 13, 13, subsample=(1, 1), border_mode='same')
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
# Channel 1 - Convolution Net Layer 5
x = conv2D_lrn2d(x, 256, 13, 13, subsample=(1, 1), border_mode='same')
x = MaxPooling2D(strides=(2, 2),
pool_size=(2, 2),
dim_ordering=DIM_ORDERING)(x)
x = ZeroPadding2D(padding=(1, 1), dim_ordering=DIM_ORDERING)(x)
# Channel 1 - Cov Net Layer 7
x = Flatten()(x)
x = Dense(4096, activation='relu')(x)
x = Dropout(DROPOUT)(x)
# Channel 1 - Cov Net Layer 8
x = Dense(4096, activation='relu')(x)
x = Dropout(DROPOUT)(x)
# Final Channel - Cov Net 9
x = Dense(output_dim=NB_CLASS, activation='softmax')(x)
return x, img_input, CONCAT_AXIS, INP_SHAPE, DIM_ORDERING