def construct_inceptionv4onfire(x,y, training=False, enable_batch_norm=True):
network = input_data(shape=[None, y, x, 3])
#stem of inceptionV4
conv1_3_3 = conv_2d(network,32,3,strides=2,activation='relu',name='conv1_3_3_s2',padding='valid')
conv2_3_3 = conv_2d(conv1_3_3,32,3,activation='relu',name='conv2_3_3')
conv3_3_3 = conv_2d(conv2_3_3,64,3,activation='relu',name='conv3_3_3')
b_conv_1_pool = max_pool_2d(conv3_3_3,kernel_size=3,strides=2,padding='valid',name='b_conv_1_pool')
if enable_batch_norm:
b_conv_1_pool = batch_normalization(b_conv_1_pool)
b_conv_1_conv = conv_2d(conv3_3_3,96,3,strides=2,padding='valid',activation='relu',name='b_conv_1_conv')
b_conv_1 = merge([b_conv_1_conv,b_conv_1_pool],mode='concat',axis=3)
b_conv4_1_1 = conv_2d(b_conv_1,64,1,activation='relu',name='conv4_3_3')
b_conv4_3_3 = conv_2d(b_conv4_1_1,96,3,padding='valid',activation='relu',name='conv5_3_3')
b_conv4_1_1_reduce = conv_2d(b_conv_1,64,1,activation='relu',name='b_conv4_1_1_reduce')
b_conv4_1_7 = conv_2d(b_conv4_1_1_reduce,64,[1,7],activation='relu',name='b_conv4_1_7')
b_conv4_7_1 = conv_2d(b_conv4_1_7,64,[7,1],activation='relu',name='b_conv4_7_1')
b_conv4_3_3_v = conv_2d(b_conv4_7_1,96,3,padding='valid',name='b_conv4_3_3_v')
b_conv_4 = merge([b_conv4_3_3_v, b_conv4_3_3],mode='concat',axis=3)
b_conv5_3_3 = conv_2d(b_conv_4,192,3,padding='valid',activation='relu',name='b_conv5_3_3',strides=2)
b_pool5_3_3 = max_pool_2d(b_conv_4,kernel_size=3,padding='valid',strides=2,name='b_pool5_3_3')
if enable_batch_norm:
b_pool5_3_3 = batch_normalization(b_pool5_3_3)
b_conv_5 = merge([b_conv5_3_3,b_pool5_3_3],mode='concat',axis=3)
net = b_conv_5
# inceptionV4 modules
net=inception_block_a(net)
net=inception_block_b(net)
net=inception_block_c(net)
pool5_7_7=global_avg_pool(net)
if(training):
pool5_7_7=dropout(pool5_7_7,0.4)
loss = fully_connected(pool5_7_7, 2,activation='softmax')
if(training):
network = regression(loss, optimizer='rmsprop',
loss='categorical_crossentropy',
learning_rate=0.001)
else:
network=loss
model = tflearn.DNN(network, checkpoint_path='inceptionv4onfire',
max_checkpoints=1, tensorboard_verbose=0)
return model
def _bottleneck(input_net, filters, kernel, t, s, r=False): """Bottleneck This function defines a basic bottleneck structure. Parameters ---------- input_net: Tensor, input tensor of conv layer. filters: Integer, the dimensionality of the output space. kernel: An integer or tuple/list of 2 integers, specifying the width and height of the 2D convolution window. t: Integer, expansion factor. t is always applied to the input size. s: An integer or tuple/list of 2 integers,specifying the strides of the convolution along the width and height.Can be a single integer to specify the same value for all spatial dimensions. r: Boolean, Whether to use the residuals. # Returns Output tensor. """ t_channel = input_net.shape[3] * t # channel expansion net = _conv_block(input_net, t_channel, (1, 1), (1, 1)) net = grouped_conv_2d(net, channel_multiplier=1, filter_size=kernel, strides=(s, s), padding='same', activation='relu6', weights_init='xavier') net = batch_normalization(net) net = conv_2d(net, filters, (1, 1), strides=(1, 1), padding='same') net = batch_normalization(net) if r: net = merge([net, input_net], 'elemwise_sum') return net
def cnn(self):
network = input_data(shape=[None, self.max_document_length],
name='input')
network = tflearn.embedding(network, input_dim=1000000, output_dim=128)
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.8)
network = fully_connected(network, 2, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
model = tflearn.DNN(network, tensorboard_verbose=0)
return model
def transform_embedded_sequences(self, embedded_sequences):
drop_1, drop_2 = self.dropout_rates
net = dropout(embedded_sequences, drop_1)
conv_blocks = []
for sz in self.filter_sizes:
conv = conv_1d(net,
nb_filter=self.num_filters,
filter_size=sz,
padding="valid",
activation="relu",
regularizer="L2")
conv_blocks.append(conv)
net = merge(conv_blocks, mode='concat',
axis=1) if len(conv_blocks) > 1 else conv_blocks[0]
net = tf.expand_dims(net, 2)
net = global_max_pool(net)
net = dropout(net, drop_2)
model_output = fully_connected(net,
self.class_count,
activation="softmax")
return model_output
def conv_model(network):
branch1 = conv_1d(network,
200,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
200,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
200,
5,
padding='valid',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
return network
def get_network():
# input images
network = input_data(shape=[None,33,33,4], name='input')
# local pathway [conv + pool + norm]
#path1 = conv_2d(network, 32, 7, activation='relu', regularizer="L2", padding='valid')
path1 = conv_2d(network, 64, 7, activation='relu', regularizer="L2", padding='valid')
path1 = max_pool_2d(path1, 4, 1, padding='valid')
path1 = dropout(path1, 0.5)
#path1 = local_response_normalization(path1)
#path1 = conv_2d(path1, 32, 3, activation='relu', regularizer="L2", padding='valid')
path1 = conv_2d(path1, 64, 3, activation='relu', regularizer="L2", padding='valid')
path1 = max_pool_2d(path1, 2, 1, padding='valid')
path1 = dropout(path1, 0.5)
#path1 = local_response_normalization(path1)
# global pathway
#path2 = conv_2d(network, 80, 13, activation='relu', regularizer="L2", padding='valid')
path2 = conv_2d(network, 160, 13, activation='relu', regularizer="L2", padding='valid')
path2 = dropout(path2, 0.5)
#path2 = local_response_normalization(path2)
network = merge([path1,path2],'concat',axis=3)
network = conv_2d(network, 5, 21, activation='relu', regularizer="L2")
network = flatten(network, name="flatten")
# softmax + output layers
network = fully_connected(network, 5, activation='softmax', name='soft')
network = regression(network, optimizer='adam', learning_rate=0.00005, # 0.0001
loss='categorical_crossentropy', name='target', batch_size=500)
return network
def get_cnn_model(max_len, volcab_size):
# Building convolutional network
network = tflearn.input_data(shape=[None, max_len], name='input')
network = tflearn.embedding(network, input_dim=volcab_size, output_dim=64)
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
network = fully_connected(network, 3, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
model = tflearn.DNN(network, tensorboard_verbose=0)
return model
def conv_reflectance_neural_network_model5(
n_actions=3,
name="conv_ref4",
shape=None,
imconstpar=constantParametersImage(),
netconstpar=constantParametersNetwork()):
convnet = input_data(
shape=[None, imconstpar.height, imconstpar.width, imconstpar.channels],
name='image_input')
reflectance = tflearn.input_data(shape=[None, 6], name="reflectance_input")
convnet = conv_2d(convnet,
32,
8,
strides=4,
activation='prelu',
name='conv1')
#convnet = max_pool_2d(convnet, 2)
convnet = conv_2d(convnet,
64,
4,
strides=2,
activation='prelu',
name='conv2')
#convnet = max_pool_2d(convnet, 2)
convnet = conv_2d(convnet,
64,
3,
strides=1,
activation='prelu',
name='conv3')
#convnet = max_pool_2d(convnet, 2)
convnet = fully_connected(convnet, 128, activation='softmax', name='fc1')
convnet = dropout(convnet, .8)
convnet = fully_connected(convnet, 3, activation='sigmoid', name='fc2')
merged_net = merge([convnet, reflectance], mode='concat', axis=1)
#convnet = regression(convnet, optimizer='SGD', loss='categorical_crossentropy', name='targets')
merged_net = fully_connected(merged_net,
12,
activation='softmax',
name='m_fc1')
merged_net = fully_connected(merged_net,
n_actions,
activation='linear',
name='m_fc2')
merged_net = regression(merged_net,
optimizer='adam',
learning_rate=netconstpar.l_rate,
loss='mean_square',
name='targets')
model = tflearn.DNN(merged_net)
return model, name
def model_M_bn_cnn_1_1_filter(Input, tt_ll, training):
branch0 = M_cnn(Input=Input, num=53, training=training, tt_ll=tt_ll)
branch1 = M_cnn(Input=Input, num=43, training=training, tt_ll=tt_ll)
branch2 = M_cnn(Input=Input, num=33, training=training, tt_ll=tt_ll)
branch3 = M_cnn(Input=Input, num=23, training=training, tt_ll=tt_ll)
branch4 = M_cnn(Input=Input, num=13, training=training, tt_ll=tt_ll)
branch5 = M_cnn(Input=Input, num=11, training=training, tt_ll=tt_ll)
branch6 = M_cnn(Input=Input, num=9, training=training, tt_ll=tt_ll)
branch7 = M_cnn(Input=Input, num=7, training=training, tt_ll=tt_ll)
branch8 = M_cnn(Input=Input, num=5, training=training, tt_ll=tt_ll)
branch9 = M_cnn(Input=Input, num=3, training=training, tt_ll=tt_ll)
#TensorShape([Dimension(None), Dimension(300), Dimension(32)])
network_out = merge([
branch0, branch1, branch2, branch3, branch4, branch5, branch6, branch7,
branch8, branch9
],
mode='concat',
axis=2)
#TensorShape([Dimension(None), Dimension(300), Dimension(320)])
#network_feature_1 = fully_connected(network_out, 1280, activation='relu')
#network_feature = fully_connected(network_feature_1, 128, activation='relu')
network_feature = dropout(network_out, 0.9)
Output = fully_connected(network_feature, 2, activation='relu')
return Output
def do_cnn(trainX, trainY,testX, testY):
global n_words
# Data preprocessing
# Sequence padding
trainX = pad_sequences(trainX, maxlen=MAX_DOCUMENT_LENGTH, value=0.)
testX = pad_sequences(testX, maxlen=MAX_DOCUMENT_LENGTH, value=0.)
# Converting labels to binary vectors
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# Building convolutional network
network = input_data(shape=[None, MAX_DOCUMENT_LENGTH], name='input')
network = tflearn.embedding(network, input_dim=n_words+1, output_dim=128)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX, trainY, n_epoch = 20, shuffle=True, validation_set=(testX, testY), show_metric=True, batch_size=32)
def do_cnn_word2vec(trainX, testX, trainY, testY):
global max_features
print "CNN and word2vec"
#trainX = pad_sequences(trainX, maxlen=max_document_length, value=-1.)
#testX = pad_sequences(testX, maxlen=max_document_length, value=0.)
# Converting labels to binary vectors
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# Building convolutional network
network = input_data(shape=[None,max_features], name='input')
network = tflearn.embedding(network, input_dim=1000000, output_dim=128)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.8)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX, trainY,
n_epoch=5, shuffle=True, validation_set=(testX, testY),
show_metric=True, batch_size=2,run_id="spam")
def build():
network = input_data([None, Meta.max_string_len])
network = embedding(network, input_dim=Meta.max_one_hot, output_dim=128)
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = dropout(network, 0.5)
network = lstm(network, 128)
# network = fully_connected(network, 20)
network = fully_connected(network, 2, activation='softmax')
network = tflearn.regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy')
model = tflearn.DNN(network, tensorboard_verbose=0)
return model
def build_network(optimizer):
net = input_data(shape=[None, length], name='input')
net = tflearn.embedding(net,
input_dim=caes_ngram_data.dims,
output_dim=128)
branch1 = conv_1d(net, 128, 3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(net, 128, 4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(net, 128, 5,
padding='valid',
activation='relu',
regularizer="L2")
net = merge([branch1, branch2, branch3], mode='concat', axis=1)
net = tf.expand_dims(net, 2)
net = global_max_pool(net)
net = dropout(net, 0.33)
net = fully_connected(net, 6, activation='softmax')
net = regression(net,
optimizer=optimizer,
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
return net
def do_cnn_doc2vec(trainX, testX, trainY, testY):
global max_features
print "CNN and doc2vec"
#trainX = pad_sequences(trainX, maxlen=max_features, value=0.)
#testX = pad_sequences(testX, maxlen=max_features, value=0.)
# Converting labels to binary vectors
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# Building convolutional network
network = input_data(shape=[None,max_features], name='input')
network = tflearn.embedding(network, input_dim=1000000, output_dim=128,validate_indices=False)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.8)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX, trainY,
n_epoch=5, shuffle=True, validation_set=(testX, testY),
show_metric=True, batch_size=100,run_id="review")
def cnn_3_filters(trainX, trainY, valX, valY, testX, input_weights):
'''
A CNN with three convolutional layers as in Kim Yoon (Convolutional Neural Networks for Sentence Classification)
'''
# Building convolutional network
network = input_data(shape=[None, MAX_LENGHT], name='input')
network = tflearn.embedding(network,
input_dim=input_weights.shape[0],
output_dim=input_weights.shape[1],
trainable=True,
name="EmbeddingLayer")
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
network = fully_connected(network, 12, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=1)
# Add embedding weights into the embedding layer
embeddingWeights = tflearn.get_layer_variables_by_name("EmbeddingLayer")[0]
model.set_weights(embeddingWeights, input_weights)
print("Start trianing CNN...")
model.fit(trainX,
trainY,
n_epoch=NB_EPOCHS,
validation_set=(valX, valY),
shuffle=True,
show_metric=True,
batch_size=32)
y_result = model.predict(testX)
return y_result
def get_network(): network = input_data(shape=[None, LEN_HISTORY], name="inp") network = reshape(network, [tf.shape(network)[0], LEN_HISTORY, 1], name="added_inp_dim") network = lstm(network, LEN_HISTORY, name="hidden", return_seq=True) network = [fully_connected(ts, 1, activation='softmax') for ts in network] network = merge(network, "concat", axis=1, name="concat_merge") network = network / tf.expand_dims(tf.reduce_mean(network, axis=1), 1, name="normed") network = regression(network, optimizer="sgd", loss=revenue) return network
def inception(self,
network,
filter_num,
activation='relu',
regularizer=None,
strides=1,
trainable=True):
network1 = conv_2d(network,
int(filter_num / 3),
1,
activation='relu',
strides=1,
regularizer=regularizer,
trainable=trainable)
network1 = conv_2d(network1,
filter_num,
5,
activation='relu',
strides=strides,
regularizer=regularizer,
trainable=trainable)
network2 = conv_2d(network,
int(filter_num / 3),
1,
activation='relu',
strides=1,
regularizer=regularizer,
trainable=trainable)
network2 = conv_2d(network2,
filter_num,
3,
activation='relu',
strides=strides,
regularizer=regularizer,
trainable=trainable)
network3 = conv_2d(network,
int(filter_num / 3),
1,
activation='relu',
strides=1,
regularizer=regularizer,
trainable=trainable)
network3 = conv_2d(network3,
filter_num,
2,
activation='relu',
strides=strides,
regularizer=regularizer,
trainable=trainable)
# network4 = max_pool_2d(network, kernel_size=2, strides=strides)
# network5 = max_pool_2d(network, kernel_size=3, strides=strides)
# network6 = max_pool_2d(network, kernel_size=5, strides=strides)
network = merge([network1, network2, network3], mode='concat', axis=3)
return network
def inception_block_a(input_a):
inception_a_conv1_1_1 = conv_2d(input_a,
96,
1,
activation='relu',
name='inception_a_conv1_1_1')
inception_a_conv1_3_3_reduce = conv_2d(input_a,
64,
1,
activation='relu',
name='inception_a_conv1_3_3_reduce')
inception_a_conv1_3_3 = conv_2d(inception_a_conv1_3_3_reduce,
96,
3,
activation='relu',
name='inception_a_conv1_3_3')
inception_a_conv2_3_3_reduce = conv_2d(input_a,
64,
1,
activation='relu',
name='inception_a_conv2_3_3_reduce')
inception_a_conv2_3_3_sym_1 = conv_2d(inception_a_conv2_3_3_reduce,
96,
3,
activation='relu',
name='inception_a_conv2_3_3')
inception_a_conv2_3_3 = conv_2d(inception_a_conv2_3_3_sym_1,
96,
3,
activation='relu',
name='inception_a_conv2_3_3')
inception_a_pool = avg_pool_2d(input_a,
kernel_size=3,
name='inception_a_pool',
strides=1)
inception_a_pool_1_1 = conv_2d(inception_a_pool,
96,
1,
activation='relu',
name='inception_a_pool_1_1')
# merge inception_a
inception_a = merge([
inception_a_conv1_1_1, inception_a_conv1_3_3, inception_a_conv2_3_3,
inception_a_pool_1_1
],
mode='concat',
axis=3)
return inception_a
def create_model_for_multiple_input(self): # main cnn model
network1 = input_data(
shape=[None, self.title_mat_size, self.title_mat_size, 1])
network2 = input_data(
shape=[None, self.author_mat_size, self.author_mat_size, 1])
activation = 'relu'
# activation = 'tanh'
# activation = 'sigmoid'
network1 = conv_2d(network1,
8,
3,
activation=activation,
regularizer="L2")
# network = max_pool_2d(network, 2)
# network = local_response_normalization(network)
network1 = conv_2d(network1,
16,
2,
activation=activation,
regularizer="L2")
# network = max_pool_2d(network, 2)
# network = local_response_normalization(network)
network1 = fully_connected(network1, 512, activation=activation)
network2 = conv_2d(network2,
8,
2,
activation=activation,
regularizer='L2')
network2 = fully_connected(network2, 512, activation=activation)
network = merge([network1, network2], 'concat')
network = dropout(network, 0.5)
network = fully_connected(network, 2, activation='softmax')
acc = Accuracy(name="Accuracy")
adagrad = tflearn.AdaGrad(learning_rate=0.002,
initial_accumulator_value=0.01)
network = regression(network,
optimizer=adagrad,
loss='categorical_crossentropy',
learning_rate=0.002,
metric=acc)
tmp_dir = join(settings.TRAIN_DIR, 'tmp')
# util.mkdir(tmp_dir)
model = tflearn.DNN(network,
checkpoint_path=join(tmp_dir, 'paper_similarity'),
max_checkpoints=10,
tensorboard_verbose=3,
tensorboard_dir=join(tmp_dir, 'tflearn_logs'))
return model
def do_cnn(x,y):
print("start CNN......")
global max_document_length
print("CNN and tf")
trainX, testX, trainY, testY = train_test_split(x, y, test_size=0.4, random_state=0)
y_test=testY
# 训练、测试数据进行填充和转换，不到最大长度的数据填充0
trainX = pad_sequences(trainX, maxlen=max_document_length, value=0.)
testX = pad_sequences(testX, maxlen=max_document_length, value=0.)
# Converting labels to binary vectors
# 二分类问题，把标记数据二值化
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# Building convolutional network
network = input_data(shape=[None,max_document_length], name='input')
# 三个数量为128，长度分别为3，4，5的一维卷积函数处理数据
network = tflearn.embedding(network, input_dim=1000000, output_dim=128)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.8)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
# 实例化CNN对象并进行训练数据，训练5轮
model = tflearn.DNN(network, tensorboard_verbose=0)
if not os.path.exists(pkl_file):
model.fit(trainX, trainY,
n_epoch=5, shuffle=True, validation_set=0.1,
show_metric=True, batch_size=100,run_id="webshell")
model.save(pkl_file)
else:
model.load(pkl_file)
y_predict_list=model.predict(testX)
y_predict=[]
for i in y_predict_list:
print(i[0])
if i[0] > 0.5:
y_predict.append(0)
else:
y_predict.append(1)
print('y_predict_list:')
print(y_predict_list)
print('y_predict:')
print(y_predict)
#print y_test
do_metrics(y_test, y_predict)
def add_merge_layer(self,
prev_layer_names,
merge_mode,
exclude_from_path=True,
**kwargs):
prev = [self._layers[name]['layer'] for name in prev_layer_names]
layer = {
"layer": merge.merge(prev, merge_mode, **kwargs),
"type": "Merge",
"categories": ["hidden"]
}
return self._add_layer(layer, exclude_from_path)
def do_cnn(x,y):
global max_document_length
print "CNN and tf"
trainX, testX, trainY, testY = train_test_split(x, y, test_size=0.4, random_state=0)
y_test=testY
trainX = pad_sequences(trainX, maxlen=max_document_length, value=0.)
testX = pad_sequences(testX, maxlen=max_document_length, value=0.)
# Converting labels to binary vectors
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# Building convolutional network
network = input_data(shape=[None,max_document_length], name='input')
network = tflearn.embedding(network, input_dim=1000000, output_dim=128)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.8)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
model = tflearn.DNN(network, tensorboard_verbose=0)
#if not os.path.exists(pkl_file):
# Training
model.fit(trainX, trainY,
n_epoch=5, shuffle=True, validation_set=0.1,
show_metric=True, batch_size=100,run_id="webshell")
# model.save(pkl_file)
#else:
# model.load(pkl_file)
y_predict_list=model.predict(testX)
#y_predict = list(model.predict(testX,as_iterable=True))
y_predict=[]
for i in y_predict_list:
print i[0]
if i[0] > 0.5:
y_predict.append(0)
else:
y_predict.append(1)
print 'y_predict_list:'
print y_predict_list
print 'y_predict:'
print y_predict
#print y_test
do_metrics(y_test, y_predict)
def create_cnn_architecture_two_layers(
self,
model_name,
outputDim=300,
number_of_filters=60,
filterSize=[3, 4],
padding='same',
activation_function_convLayer='relu',
regularizer='L2',
dropouts=0.5,
activation_function_fc='softmax',
optimizer='adam',
learning_rate=0.001,
loss_function='categorical_crossentropy'):
if len(filterSize) == 0:
filterSize = [3, 4]
""" Define input shape and create word embedding """
self.cnn_model = input_data(shape=[None, self.max_words], name='input')
self.cnn_model = tflearn.embedding(
self.cnn_model,
input_dim=len(self.vocabProcessor.vocabulary_),
output_dim=outputDim)
""" Add three/two convolutional layer. Set number of filters and filter sizes and then merge together """
conv1 = conv_1d(self.cnn_model,
nb_filter=number_of_filters,
filter_size=filterSize[0],
padding=padding,
activation=activation_function_convLayer,
regularizer=regularizer)
conv2 = conv_1d(self.cnn_model,
nb_filter=number_of_filters,
filter_size=filterSize[1],
padding=padding,
activation=activation_function_convLayer,
regularizer=regularizer)
#conv3 = conv_1d(cnn_model, nb_filter = 128, filter_size = 5, padding = 'same',
# activation = 'relu', regularizer = 'L2')
self.cnn_model = merge([conv1, conv2], mode='concat', axis=1)
""" Expand one dimension to fit the max_pooling layer """
self.cnn_model = tf.expand_dims(self.cnn_model, 1)
self.cnn_model = global_max_pool(self.cnn_model)
""" Instantiate dropout layer and specify dropout parameter """
self.cnn_model = dropout(self.cnn_model, dropouts)
""" Instantiate fully connected layer and regression layer. """
self.cnn_model = fully_connected(self.cnn_model,
self.number_of_classes,
activation=activation_function_fc)
self.cnn_model = regression(self.cnn_model,
optimizer=optimizer,
learning_rate=learning_rate,
loss=loss_function,
name='models/' + model_name)
def block17(net, scale=1.0, activation="relu"):
tower_conv = relu(
batch_normalization(
conv_2d(net,
192,
1,
bias=False,
activation=None,
name='Conv2d_1x1')))
tower_conv_1_0 = relu(
batch_normalization(
conv_2d(net,
128,
1,
bias=False,
activation=None,
name='Conv2d_0a_1x1')))
tower_conv_1_1 = relu(
batch_normalization(
conv_2d(tower_conv_1_0,
160, [1, 7],
bias=False,
activation=None,
name='Conv2d_0b_1x7')))
tower_conv_1_2 = relu(
batch_normalization(
conv_2d(tower_conv_1_1,
192, [7, 1],
bias=False,
activation=None,
name='Conv2d_0c_7x1')))
tower_mixed = merge([tower_conv, tower_conv_1_2],
mode='concat',
axis=3)
tower_out = relu(
batch_normalization(
conv_2d(tower_mixed,
net.get_shape()[3],
1,
bias=False,
activation=None,
name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
def build_fcn_all(network):
#Pool1
conv1 = conv_2d(network, 8, 7, activation='relu')
pool1 = max_pool_2d(conv1, 2)
#Pool2
conv2 = conv_2d(pool1, 16, 5, activation='relu')
pool2 = max_pool_2d(conv2, 2)
#Pool3
conv3 = conv_2d(pool2, 32, 5, activation='relu')
pool3 = max_pool_2d(conv3, 2) # output 8x_downsampled
#Pool4
conv4 = conv_2d(pool3, 64, 3, activation='relu')
pool4 = max_pool_2d(conv4, 2) # output 16x_downsampled
#start FCN-32s -----------------------------------
#Pool5
conv5 = conv_2d(pool4, 128, 3, activation='relu')
pool5 = max_pool_2d(conv5, 2)
#Conv6-7
conv6 = conv_2d(pool5, 128, 3, activation='relu')
conv7 = conv_2d(conv6, 128, 3, activation='relu') # output 32x_downsampled
#end FCN-32s -----------------------------------
##start FCN-16s -----------------------------------
#network_32_UP2 = upsample_2d(network_32, 2)
#network_16 = merge([network_32_UP2, network_4], mode='concat', axis=3)
#network_16 = conv_2d(network_16, 3, 128, activation='relu') # output 16x_downsampled
##end FCN-16s -----------------------------------
##start FCN-8s -----------------------------------
#network_32_UP4 = upsample_2d(network_32, 4)
#network_16_UP2 = upsample_2d(network_16, 2)
#network_3_UP8 = upsample_2d(network_3, 8)
pool4_x2 = upsample_2d(pool4, 2)
conv7_x4 = upsample_2d(conv7, 4)
#network_8 = merge([network_32_UP4, network_4_UP2, network_3], mode='concat', axis=3)
fcn_8s = merge([pool3, pool4_x2, conv7_x4], mode='concat', axis=3)
fcn_8s = conv_2d(fcn_8s, 3, 1, activation='relu')
##end FCN-8s -----------------------------------
out = conv_2d(fcn_8s, CHANNELS, 1, activation='relu')
out = upsample_2d(out, 8)
#network_8 = upscore_layer(network_8, num_classes=3, kernel_size=2, strides=8, shape=[384, 1216, 3])
network = tflearn.regression(
out,
loss='mean_square',
#loss='weak_cross_entropy_2d',
)
return network
def buildNet(self):
#necessary to allow the model to be loaded after it has been trained
tf.reset_default_graph()
#build input layer to accept 140 chars
network = input_data(shape=[None, 72], name='input')
#embedded layer
network = tflearn.embedding(network,
input_dim=len(self.vp.vocabulary_) + 2,
output_dim=128)
#create three convolutional layers
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer="L2")
#merge all incoming tensors into a single tenso
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
#expand dimensions of network to 3d tensor, as input is a 1d tensor
network = tf.expand_dims(network, 2)
#perform reduction operation over input tensor
network = global_max_pool(network)
#prevent overfitting by including dropout
network = dropout(network, 0.8)
#output layer
network = fully_connected(network, 8, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.0001,
loss='categorical_crossentropy',
name='target')
return network
def do_cnn_1d(trainX, testX, trainY, testY):
# Converting labels to binary vectors
trainY = to_categorical(trainY, nb_classes=4)
testY = to_categorical(testY, nb_classes=4)
# Building convolutional network
network = input_data(shape=[None, 1000], name='input')
network = tflearn.embedding(network,
input_dim=1000000,
output_dim=128,
validate_indices=False)
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.8)
network = fully_connected(network, 4, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX,
trainY,
n_epoch=5,
shuffle=True,
validation_set=(testX, testY),
show_metric=True,
batch_size=100,
run_id="malware")
def reduction_block_b(reduction_input_b):
reduction_b_1_1 = conv_2d(reduction_input_b,
192,
1,
activation='relu',
name='reduction_b_1_1')
reduction_b_1_3 = conv_2d(reduction_b_1_1,
192,
3,
strides=2,
padding='valid',
name='reduction_b_1_3')
reduction_b_3_3_reduce = conv_2d(reduction_input_b,
256,
filter_size=1,
activation='relu',
name='reduction_b_3_3_reduce')
reduction_b_3_3_asym_1 = conv_2d(reduction_b_3_3_reduce,
256,
filter_size=[1, 7],
activation='relu',
name='reduction_b_3_3_asym_1')
reduction_b_3_3_asym_2 = conv_2d(reduction_b_3_3_asym_1,
320,
filter_size=[7, 1],
activation='relu',
name='reduction_b_3_3_asym_2')
reduction_b_3_3 = conv_2d(reduction_b_3_3_asym_2,
320,
3,
strides=2,
activation='relu',
padding='valid',
name='reduction_b_3_3')
reduction_b_pool = max_pool_2d(reduction_input_b,
kernel_size=3,
strides=2,
padding='valid')
# merge the reduction_b
reduction_b_output = merge(
[reduction_b_1_3, reduction_b_3_3, reduction_b_pool],
mode='concat',
axis=3)
return reduction_b_output
def build():
network = input_data([None, Meta.max_string_len])
network = embedding(network, input_dim=Meta.max_one_hot, output_dim=128)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = dropout(network, 0.5)
network = lstm(network, 128)
# network = fully_connected(network, 20)
network = fully_connected(network, 2, activation='softmax')
network = tflearn.regression(network, optimizer='adam', learning_rate=0.001, loss='categorical_crossentropy')
model = tflearn.DNN(network, tensorboard_verbose=0)
return model
def model_2(train_x, train_y, test_x, test_y, embedding_size):
# train_x = pad_sequences(train_x, maxlen=100, value=0.)
# test_x = pad_sequences(test_x, maxlen=100, value=0.)
out_dim = embedding_size # embedding size
num_cat = len(train_y[0])
network = input_data(shape=[None, len(train_x[0])], name='input')
network = tflearn.embedding(network,
input_dim=len(train_x[0]),
output_dim=out_dim) # input_dim - vocab size
branch1 = conv_1d(network,
out_dim,
3,
padding='same',
activation='relu',
regularizer="L2")
branch2 = conv_1d(network,
out_dim,
4,
padding='same',
activation='relu',
regularizer="L2")
branch3 = conv_1d(network,
out_dim,
5,
padding='same',
activation='relu',
regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
network = fully_connected(network, num_cat, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(train_x,
train_y,
n_epoch=5,
shuffle=True,
validation_set=(test_x, test_y),
show_metric=True,
batch_size=32)
return model
def reduction_block_a(reduction_input_a):
reduction_a_conv1_1_1 = conv_2d(reduction_input_a,384,3,strides=2,padding='valid',activation='relu',name='reduction_a_conv1_1_1')
reduction_a_conv2_1_1 = conv_2d(reduction_input_a,192,1,activation='relu',name='reduction_a_conv2_1_1')
reduction_a_conv2_3_3 = conv_2d(reduction_a_conv2_1_1,224,3,activation='relu',name='reduction_a_conv2_3_3')
reduction_a_conv2_3_3_s2 = conv_2d(reduction_a_conv2_3_3,256,3,strides=2,padding='valid',activation='relu',name='reduction_a_conv2_3_3_s2')
reduction_a_pool = max_pool_2d(reduction_input_a,strides=2,padding='valid',kernel_size=3,name='reduction_a_pool')
# merge reduction_a
reduction_a = merge([reduction_a_conv1_1_1,reduction_a_conv2_3_3_s2,reduction_a_pool],mode='concat',axis=3)
return reduction_a
def inception_block_c(input_c):
inception_c_1_1 = conv_2d(input_c, 256, 1, activation='relu', name='inception_c_1_1')
inception_c_3_3_reduce = conv_2d(input_c, 384, filter_size=1, activation='relu', name='inception_c_3_3_reduce')
inception_c_3_3_asym_1 = conv_2d(inception_c_3_3_reduce, 256, filter_size=[1,3], activation='relu',name='inception_c_3_3_asym_1')
inception_c_3_3_asym_2 = conv_2d(inception_c_3_3_reduce, 256, filter_size=[3,1], activation='relu',name='inception_c_3_3_asym_2')
inception_c_3_3=merge([inception_c_3_3_asym_1,inception_c_3_3_asym_2],mode='concat',axis=3)
inception_c_5_5_reduce = conv_2d(input_c, 384, filter_size=1, activation='relu', name = 'inception_c_5_5_reduce')
inception_c_5_5_asym_1 = conv_2d(inception_c_5_5_reduce, 448, filter_size=[1,3], name = 'inception_c_5_5_asym_1')
inception_c_5_5_asym_2 = conv_2d(inception_c_5_5_asym_1, 512, filter_size=[3,1], activation='relu',name='inception_c_5_5_asym_2')
inception_c_5_5_asym_3 = conv_2d(inception_c_5_5_asym_2, 256, filter_size=[1,3], activation='relu',name='inception_c_5_5_asym_3')
inception_c_5_5_asym_4 = conv_2d(inception_c_5_5_asym_2, 256, filter_size=[3,1], activation='relu',name='inception_c_5_5_asym_4')
inception_c_5_5=merge([inception_c_5_5_asym_4,inception_c_5_5_asym_3],mode='concat',axis=3)
inception_c_pool = avg_pool_2d(input_c, kernel_size=3, strides=1 )
inception_c_pool_1_1 = conv_2d(inception_c_pool, 256, filter_size=1, activation='relu', name='inception_c_pool_1_1')
# merge the inception_c
inception_c_output = merge([inception_c_1_1, inception_c_3_3, inception_c_5_5, inception_c_pool_1_1], mode='concat', axis=3)
return inception_c_output
def nlp_cnn(trainX, trainY, testX, testY):
# pad the sequence
trainX = pad_sequences(trainX, maxlen=100, value=0.)
testX = pad_sequences(testX, maxlen=100, value=0.)
# one_hot encoding
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# build an embedding
network = input_data(shape=[None, 100], name='input')
network = tflearn.embedding(network, input_dim=10000, output_dim=128)
# build an convnet
branch1 = conv_1d(network,
128,
3,
padding='valid',
activation='relu',
regularizer='L2')
branch2 = conv_1d(network,
128,
4,
padding='valid',
activation='relu',
regularizer='L2')
branch3 = conv_1d(network,
128,
5,
padding='valid',
activation='relu',
regularizer='L2')
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
network = fully_connected(network, 2, activation='softmax')
network = regression(network,
optimizer='adam',
learning_rate=0.001,
loss='categorical_crossentropy',
name='target')
# training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX,
trainY,
n_epoch=5,
shuffle=True,
validation_set=(testX, testY),
show_metric=True,
batch_size=32)
def block8(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 224, [1,3], bias=False, activation=None, name='Conv2d_0b_1x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 256, [3,1], bias=False, name='Conv2d_0c_3x1')))
tower_mixed = merge([tower_conv,tower_conv1_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
def do_cnn_word2vec_2d_345(trainX, testX, trainY, testY):
global max_features
global max_document_length
print "CNN and word2vec_2d_345"
y_test = testY
trainY = to_categorical(trainY, nb_classes=2)
testY = to_categorical(testY, nb_classes=2)
# Building convolutional network
network = input_data(shape=[None,max_document_length,max_features,1], name='input')
network = tflearn.embedding(network, input_dim=1, output_dim=128,validate_indices=False)
branch1 = conv_2d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_2d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_2d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool_2d(network)
network = dropout(network, 0.8)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX, trainY,
n_epoch=5, shuffle=True, validation_set=(testX, testY),
show_metric=True, batch_size=100,run_id="sms")
y_predict_list = model.predict(testX)
print y_predict_list
y_predict = []
for i in y_predict_list:
print i[0]
if i[0] > 0.5:
y_predict.append(0)
else:
y_predict.append(1)
print(classification_report(y_test, y_predict))
print metrics.confusion_matrix(y_test, y_predict)
def _model5():
global yTest, img_aug
tf.reset_default_graph()
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()
def block35(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 32, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 32, 1, bias=False, activation=None,name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 32, 3, bias=False, activation=None,name='Conv2d_0b_3x3')))
tower_conv2_0 = relu(batch_normalization(conv_2d(net, 32, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2_0, 48,3, bias=False, activation=None, name='Conv2d_0b_3x3')))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 64,3, bias=False, activation=None, name='Conv2d_0c_3x3')))
tower_mixed = merge([tower_conv, tower_conv1_1, tower_conv2_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
def block17(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv_1_0 = relu(batch_normalization(conv_2d(net, 128, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv_1_1 = relu(batch_normalization(conv_2d(tower_conv_1_0, 160,[1,7], bias=False, activation=None,name='Conv2d_0b_1x7')))
tower_conv_1_2 = relu(batch_normalization(conv_2d(tower_conv_1_1, 192, [7,1], bias=False, activation=None,name='Conv2d_0c_7x1')))
tower_mixed = merge([tower_conv,tower_conv_1_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
def block8(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 224, [1,3], bias=False, activation=None, name='Conv2d_0b_1x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 256, [3,1], bias=False, name='Conv2d_0c_3x1')))
tower_mixed = merge([tower_conv,tower_conv1_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
num_classes = len(Y[0])
dropout_keep_prob = 0.8
network = input_data(shape=[None, inputSize, inputSize, dim],
name='input',
data_preprocessing=img_prep,
data_augmentation=img_aug)
conv1a_3_3 = relu(batch_normalization(conv_2d(network, 32, 3, strides=2, bias=False, padding='VALID',activation=None,name='Conv2d_1a_3x3')))
conv2a_3_3 = relu(batch_normalization(conv_2d(conv1a_3_3, 32, 3, bias=False, padding='VALID',activation=None, name='Conv2d_2a_3x3')))
conv2b_3_3 = relu(batch_normalization(conv_2d(conv2a_3_3, 64, 3, bias=False, activation=None, name='Conv2d_2b_3x3')))
maxpool3a_3_3 = max_pool_2d(conv2b_3_3, 3, strides=2, padding='VALID', name='MaxPool_3a_3x3')
conv3b_1_1 = relu(batch_normalization(conv_2d(maxpool3a_3_3, 80, 1, bias=False, padding='VALID',activation=None, name='Conv2d_3b_1x1')))
conv4a_3_3 = relu(batch_normalization(conv_2d(conv3b_1_1, 192, 3, bias=False, padding='VALID',activation=None, name='Conv2d_4a_3x3')))
maxpool5a_3_3 = max_pool_2d(conv4a_3_3, 3, strides=2, padding='VALID', name='MaxPool_5a_3x3')
tower_conv = relu(batch_normalization(conv_2d(maxpool5a_3_3, 96, 1, bias=False, activation=None, name='Conv2d_5b_b0_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(maxpool5a_3_3, 48, 1, bias=False, activation=None, name='Conv2d_5b_b1_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 64, 5, bias=False, activation=None, name='Conv2d_5b_b1_0b_5x5')))
tower_conv2_0 = relu(batch_normalization(conv_2d(maxpool5a_3_3, 64, 1, bias=False, activation=None, name='Conv2d_5b_b2_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2_0, 96, 3, bias=False, activation=None, name='Conv2d_5b_b2_0b_3x3')))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 96, 3, bias=False, activation=None,name='Conv2d_5b_b2_0c_3x3')))
tower_pool3_0 = avg_pool_2d(maxpool5a_3_3, 3, strides=1, padding='same', name='AvgPool_5b_b3_0a_3x3')
tower_conv3_1 = relu(batch_normalization(conv_2d(tower_pool3_0, 64, 1, bias=False, activation=None,name='Conv2d_5b_b3_0b_1x1')))
tower_5b_out = merge([tower_conv, tower_conv1_1, tower_conv2_2, tower_conv3_1], mode='concat', axis=3)
net = repeat(tower_5b_out, 10, block35, scale=0.17)
'''
tower_conv = relu(batch_normalization(conv_2d(net, 384, 3, bias=False, strides=2,activation=None, padding='VALID', name='Conv2d_6a_b0_0a_3x3')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 256, 3, bias=False, activation=None, name='Conv2d_6a_b1_0b_3x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 384, 3, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_6a_b1_0c_3x3')))
tower_pool = max_pool_2d(net, 3, strides=2, padding='VALID',name='MaxPool_1a_3x3')
net = merge([tower_conv, tower_conv1_2, tower_pool], mode='concat', axis=3)
net = repeat(net, 20, block17, scale=0.1)
tower_conv = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv0_1 = relu(batch_normalization(conv_2d(tower_conv, 384, 3, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1,288,3, bias=False, strides=2, padding='VALID',activation=None, name='COnv2d_1a_3x3')))
tower_conv2 = relu(batch_normalization(conv_2d(net, 256,1, bias=False, activation=None,name='Conv2d_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2, 288,3, bias=False, name='Conv2d_0b_3x3',activation=None)))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 320, 3, bias=False, strides=2, padding='VALID',activation=None, name='Conv2d_1a_3x3')))
tower_pool = max_pool_2d(net, 3, strides=2, padding='VALID', name='MaxPool_1a_3x3')
'''
tower_conv = relu(batch_normalization(conv_2d(net, 384, 1, bias=False, strides=2,activation=None, padding='VALID', name='Conv2d_6a_b0_0a_3x3')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0b_3x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 384, 1, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_6a_b1_0c_3x3')))
tower_pool = max_pool_2d(net, 1, strides=2, padding='VALID',name='MaxPool_1a_3x3')
net = merge([tower_conv, tower_conv1_2, tower_pool], mode='concat', axis=3)
net = repeat(net, 20, block17, scale=0.1)
tower_conv = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv0_1 = relu(batch_normalization(conv_2d(tower_conv, 384, 1, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1,288,1, bias=False, strides=2, padding='VALID',activation=None, name='COnv2d_1a_3x3')))
tower_conv2 = relu(batch_normalization(conv_2d(net, 256,1, bias=False, activation=None,name='Conv2d_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2, 288,1, bias=False, name='Conv2d_0b_3x3',activation=None)))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 320, 1, bias=False, strides=2, padding='VALID',activation=None, name='Conv2d_1a_3x3')))
tower_pool = max_pool_2d(net, 1, strides=2, padding='VALID', name='MaxPool_1a_3x3')
####
net = merge([tower_conv0_1, tower_conv1_1,tower_conv2_2, tower_pool], mode='concat', axis=3)
net = repeat(net, 9, block8, scale=0.2)
net = block8(net, activation=None)
net = relu(batch_normalization(conv_2d(net, 1536, 1, bias=False, activation=None, name='Conv2d_7b_1x1')))
net = avg_pool_2d(net, net.get_shape().as_list()[1:3],strides=2, padding='VALID', name='AvgPool_1a_8x8')
net = flatten(net)
net = dropout(net, dropout_keep_prob)
loss = fully_connected(net, num_classes,activation='softmax')
network = tflearn.regression(loss, optimizer='RMSprop',
loss='categorical_crossentropy',
learning_rate=0.0001)
model = tflearn.DNN(network, checkpoint_path='inception_resnet_v2',
max_checkpoints=1, tensorboard_verbose=2, tensorboard_dir="./tflearn_logs/")
model.fit(X, Y, n_epoch=epochNum, validation_set=(xTest, yTest), shuffle=True,
show_metric=True, batch_size=batchNum, snapshot_step=2000,
snapshot_epoch=False, run_id='inception_resnet_v2_oxflowers17')
if modelStore: model.save(_id + '-model.tflearn')
conv4a_3_3 = relu(batch_normalization(conv_2d(conv3b_1_1, 192, 3, bias=False, padding='VALID',activation=None, name='Conv2d_4a_3x3'))) maxpool5a_3_3 = max_pool_2d(conv4a_3_3, 3, strides=2, padding='VALID', name='MaxPool_5a_3x3') tower_conv = relu(batch_normalization(conv_2d(maxpool5a_3_3, 96, 1, bias=False, activation=None, name='Conv2d_5b_b0_1x1'))) tower_conv1_0 = relu(batch_normalization(conv_2d(maxpool5a_3_3, 48, 1, bias=False, activation=None, name='Conv2d_5b_b1_0a_1x1'))) tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 64, 5, bias=False, activation=None, name='Conv2d_5b_b1_0b_5x5'))) tower_conv2_0 = relu(batch_normalization(conv_2d(maxpool5a_3_3, 64, 1, bias=False, activation=None, name='Conv2d_5b_b2_0a_1x1'))) tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2_0, 96, 3, bias=False, activation=None, name='Conv2d_5b_b2_0b_3x3'))) tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 96, 3, bias=False, activation=None,name='Conv2d_5b_b2_0c_3x3'))) tower_pool3_0 = avg_pool_2d(maxpool5a_3_3, 3, strides=1, padding='same', name='AvgPool_5b_b3_0a_3x3') tower_conv3_1 = relu(batch_normalization(conv_2d(tower_pool3_0, 64, 1, bias=False, activation=None,name='Conv2d_5b_b3_0b_1x1'))) tower_5b_out = merge([tower_conv, tower_conv1_1, tower_conv2_2, tower_conv3_1], mode='concat', axis=3) net = repeat(tower_5b_out, 10, block35, scale=0.17) tower_conv = relu(batch_normalization(conv_2d(net, 384, 3, bias=False, strides=2,activation=None, padding='VALID', name='Conv2d_6a_b0_0a_3x3'))) tower_conv1_0 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0a_1x1'))) tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 256, 3, bias=False, activation=None, name='Conv2d_6a_b1_0b_3x3'))) tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 384, 3, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_6a_b1_0c_3x3'))) tower_pool = max_pool_2d(net, 3, strides=2, padding='VALID',name='MaxPool_1a_3x3') net = merge([tower_conv, tower_conv1_2, tower_pool], mode='concat', axis=3) net = repeat(net, 20, block17, scale=0.1) tower_conv = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_0a_1x1'))) tower_conv0_1 = relu(batch_normalization(conv_2d(tower_conv, 384, 3, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_0a_1x1'))) tower_conv1 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
def _model5():
global yTest, img_aug
tf.reset_default_graph()
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()
def block35(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 32, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 32, 1, bias=False, activation=None,name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 32, 3, bias=False, activation=None,name='Conv2d_0b_3x3')))
tower_conv2_0 = relu(batch_normalization(conv_2d(net, 32, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2_0, 48,3, bias=False, activation=None, name='Conv2d_0b_3x3')))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 64,3, bias=False, activation=None, name='Conv2d_0c_3x3')))
tower_mixed = merge([tower_conv, tower_conv1_1, tower_conv2_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
def block17(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv_1_0 = relu(batch_normalization(conv_2d(net, 128, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv_1_1 = relu(batch_normalization(conv_2d(tower_conv_1_0, 160,[1,7], bias=False, activation=None,name='Conv2d_0b_1x7')))
tower_conv_1_2 = relu(batch_normalization(conv_2d(tower_conv_1_1, 192, [7,1], bias=False, activation=None,name='Conv2d_0c_7x1')))
tower_mixed = merge([tower_conv,tower_conv_1_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
def block8(net, scale=1.0, activation="relu"):
tower_conv = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 192, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 224, [1,3], bias=False, activation=None, name='Conv2d_0b_1x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 256, [3,1], bias=False, name='Conv2d_0c_3x1')))
tower_mixed = merge([tower_conv,tower_conv1_2], mode='concat', axis=3)
tower_out = relu(batch_normalization(conv_2d(tower_mixed, net.get_shape()[3], 1, bias=False, activation=None, name='Conv2d_1x1')))
net += scale * tower_out
if activation:
if isinstance(activation, str):
net = activations.get(activation)(net)
elif hasattr(activation, '__call__'):
net = activation(net)
else:
raise ValueError("Invalid Activation.")
return net
num_classes = len(yTest[0])
dropout_keep_prob = 0.8
network = input_data(shape=[None, inputSize, inputSize, dim],
name='input',
data_preprocessing=img_prep,
data_augmentation=img_aug)
conv1a_3_3 = relu(batch_normalization(conv_2d(network, 32, 3, strides=2, bias=False, padding='VALID',activation=None,name='Conv2d_1a_3x3')))
conv2a_3_3 = relu(batch_normalization(conv_2d(conv1a_3_3, 32, 3, bias=False, padding='VALID',activation=None, name='Conv2d_2a_3x3')))
conv2b_3_3 = relu(batch_normalization(conv_2d(conv2a_3_3, 64, 3, bias=False, activation=None, name='Conv2d_2b_3x3')))
maxpool3a_3_3 = max_pool_2d(conv2b_3_3, 3, strides=2, padding='VALID', name='MaxPool_3a_3x3')
conv3b_1_1 = relu(batch_normalization(conv_2d(maxpool3a_3_3, 80, 1, bias=False, padding='VALID',activation=None, name='Conv2d_3b_1x1')))
conv4a_3_3 = relu(batch_normalization(conv_2d(conv3b_1_1, 192, 3, bias=False, padding='VALID',activation=None, name='Conv2d_4a_3x3')))
maxpool5a_3_3 = max_pool_2d(conv4a_3_3, 3, strides=2, padding='VALID', name='MaxPool_5a_3x3')
tower_conv = relu(batch_normalization(conv_2d(maxpool5a_3_3, 96, 1, bias=False, activation=None, name='Conv2d_5b_b0_1x1')))
tower_conv1_0 = relu(batch_normalization(conv_2d(maxpool5a_3_3, 48, 1, bias=False, activation=None, name='Conv2d_5b_b1_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 64, 5, bias=False, activation=None, name='Conv2d_5b_b1_0b_5x5')))
tower_conv2_0 = relu(batch_normalization(conv_2d(maxpool5a_3_3, 64, 1, bias=False, activation=None, name='Conv2d_5b_b2_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2_0, 96, 3, bias=False, activation=None, name='Conv2d_5b_b2_0b_3x3')))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 96, 3, bias=False, activation=None,name='Conv2d_5b_b2_0c_3x3')))
tower_pool3_0 = avg_pool_2d(maxpool5a_3_3, 3, strides=1, padding='same', name='AvgPool_5b_b3_0a_3x3')
tower_conv3_1 = relu(batch_normalization(conv_2d(tower_pool3_0, 64, 1, bias=False, activation=None,name='Conv2d_5b_b3_0b_1x1')))
tower_5b_out = merge([tower_conv, tower_conv1_1, tower_conv2_2, tower_conv3_1], mode='concat', axis=3)
net = repeat(tower_5b_out, 10, block35, scale=0.17)
'''
tower_conv = relu(batch_normalization(conv_2d(net, 384, 3, bias=False, strides=2,activation=None, padding='VALID', name='Conv2d_6a_b0_0a_3x3')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 256, 3, bias=False, activation=None, name='Conv2d_6a_b1_0b_3x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 384, 3, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_6a_b1_0c_3x3')))
tower_pool = max_pool_2d(net, 3, strides=2, padding='VALID',name='MaxPool_1a_3x3')
net = merge([tower_conv, tower_conv1_2, tower_pool], mode='concat', axis=3)
net = repeat(net, 20, block17, scale=0.1)
tower_conv = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv0_1 = relu(batch_normalization(conv_2d(tower_conv, 384, 3, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1,288,3, bias=False, strides=2, padding='VALID',activation=None, name='COnv2d_1a_3x3')))
tower_conv2 = relu(batch_normalization(conv_2d(net, 256,1, bias=False, activation=None,name='Conv2d_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2, 288,3, bias=False, name='Conv2d_0b_3x3',activation=None)))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 320, 3, bias=False, strides=2, padding='VALID',activation=None, name='Conv2d_1a_3x3')))
tower_pool = max_pool_2d(net, 3, strides=2, padding='VALID', name='MaxPool_1a_3x3')
'''
tower_conv = relu(batch_normalization(conv_2d(net, 384, 1, bias=False, strides=2,activation=None, padding='VALID', name='Conv2d_6a_b0_0a_3x3')))
tower_conv1_0 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1_0, 256, 1, bias=False, activation=None, name='Conv2d_6a_b1_0b_3x3')))
tower_conv1_2 = relu(batch_normalization(conv_2d(tower_conv1_1, 384, 1, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_6a_b1_0c_3x3')))
tower_pool = max_pool_2d(net, 1, strides=2, padding='VALID',name='MaxPool_1a_3x3')
net = merge([tower_conv, tower_conv1_2, tower_pool], mode='concat', axis=3)
net = repeat(net, 20, block17, scale=0.1)
tower_conv = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, activation=None, name='Conv2d_0a_1x1')))
tower_conv0_1 = relu(batch_normalization(conv_2d(tower_conv, 384, 1, bias=False, strides=2, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1 = relu(batch_normalization(conv_2d(net, 256, 1, bias=False, padding='VALID', activation=None,name='Conv2d_0a_1x1')))
tower_conv1_1 = relu(batch_normalization(conv_2d(tower_conv1,288,1, bias=False, strides=2, padding='VALID',activation=None, name='COnv2d_1a_3x3')))
tower_conv2 = relu(batch_normalization(conv_2d(net, 256,1, bias=False, activation=None,name='Conv2d_0a_1x1')))
tower_conv2_1 = relu(batch_normalization(conv_2d(tower_conv2, 288,1, bias=False, name='Conv2d_0b_3x3',activation=None)))
tower_conv2_2 = relu(batch_normalization(conv_2d(tower_conv2_1, 320, 1, bias=False, strides=2, padding='VALID',activation=None, name='Conv2d_1a_3x3')))
tower_pool = max_pool_2d(net, 1, strides=2, padding='VALID', name='MaxPool_1a_3x3')
####
net = merge([tower_conv0_1, tower_conv1_1,tower_conv2_2, tower_pool], mode='concat', axis=3)
net = repeat(net, 9, block8, scale=0.2)
net = block8(net, activation=None)
net = relu(batch_normalization(conv_2d(net, 1536, 1, bias=False, activation=None, name='Conv2d_7b_1x1')))
net = avg_pool_2d(net, net.get_shape().as_list()[1:3],strides=2, padding='VALID', name='AvgPool_1a_8x8')
net = flatten(net)
net = dropout(net, dropout_keep_prob)
loss = fully_connected(net, num_classes,activation='softmax')
network = tflearn.regression(loss, optimizer='RMSprop',
loss='categorical_crossentropy',
learning_rate=0.0001)
model = tflearn.DNN(network, checkpoint_path='inception_resnet_v2',
max_checkpoints=1, tensorboard_verbose=2, tensorboard_dir="./tflearn_logs/")
model.load(_path)
pred = model.predict(xTest)
df = pd.DataFrame(pred)
df.to_csv(_path + ".csv")
newList = pred.copy()
newList = convert2(newList)
if _CSV: makeCSV(newList)
pred = convert2(pred)
pred = convert3(pred)
yTest = convert3(yTest)
print(metrics.confusion_matrix(yTest, pred))
print(metrics.classification_report(yTest, pred))
print('Accuracy', accuracy_score(yTest, pred))
print()
if _wrFile: writeTest(pred)
def _model4():
# Taken from TFLearn examples and based on Googles Inception. DO NOT RUN!!!!
global yTest, img_aug
tf.reset_default_graph()
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()
network = input_data(shape=[None, inputSize, inputSize, dim],
name='input',
data_preprocessing=img_prep,
data_augmentation=img_aug)
conv1_7_7 = conv_2d(network, 64, 7, strides=2, activation='relu', name='conv1_7_7_s2')
pool1_3_3 = max_pool_2d(conv1_7_7, 3, strides=2)
pool1_3_3 = local_response_normalization(pool1_3_3)
conv2_3_3_reduce = conv_2d(pool1_3_3, 64, 1, activation='relu', name='conv2_3_3_reduce')
conv2_3_3 = conv_2d(conv2_3_3_reduce, 192, 3, activation='relu', name='conv2_3_3')
conv2_3_3 = local_response_normalization(conv2_3_3)
pool2_3_3 = max_pool_2d(conv2_3_3, kernel_size=3, strides=2, name='pool2_3_3_s2')
# 3a
inception_3a_1_1 = conv_2d(pool2_3_3, 64, 1, activation='relu', name='inception_3a_1_1')
inception_3a_3_3_reduce = conv_2d(pool2_3_3, 96, 1, activation='relu', name='inception_3a_3_3_reduce')
inception_3a_3_3 = conv_2d(inception_3a_3_3_reduce, 128, filter_size=3, activation='relu', name='inception_3a_3_3')
inception_3a_5_5_reduce = conv_2d(pool2_3_3, 16, filter_size=1, activation='relu', name='inception_3a_5_5_reduce')
inception_3a_5_5 = conv_2d(inception_3a_5_5_reduce, 32, filter_size=5, activation='relu', name='inception_3a_5_5')
inception_3a_pool = max_pool_2d(pool2_3_3, kernel_size=3, strides=1, name='inception_3a_pool')
inception_3a_pool_1_1 = conv_2d(inception_3a_pool, 32, filter_size=1, activation='relu', name='inception_3a_pool_1_1')
inception_3a_output = merge([inception_3a_1_1, inception_3a_3_3, inception_3a_5_5, inception_3a_pool_1_1], mode='concat', axis=3)
# 3b
inception_3b_1_1 = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_1_1')
inception_3b_3_3_reduce = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_3_3_reduce')
inception_3b_3_3 = conv_2d(inception_3b_3_3_reduce, 192, filter_size=3, activation='relu', name='inception_3b_3_3')
inception_3b_5_5_reduce = conv_2d(inception_3a_output, 32, filter_size=1, activation='relu', name='inception_3b_5_5_reduce')
inception_3b_5_5 = conv_2d(inception_3b_5_5_reduce, 96, filter_size=5, name='inception_3b_5_5')
inception_3b_pool = max_pool_2d(inception_3a_output, kernel_size=3, strides=1, name='inception_3b_pool')
inception_3b_pool_1_1 = conv_2d(inception_3b_pool, 64, filter_size=1, activation='relu', name='inception_3b_pool_1_1')
inception_3b_output = merge([inception_3b_1_1, inception_3b_3_3, inception_3b_5_5, inception_3b_pool_1_1], mode='concat', axis=3, name='inception_3b_output')
pool3_3_3 = max_pool_2d(inception_3b_output, kernel_size=3, strides=2, name='pool3_3_3')
# 4a
inception_4a_1_1 = conv_2d(pool3_3_3, 192, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4a_3_3_reduce = conv_2d(pool3_3_3, 96, filter_size=1, activation='relu', name='inception_4a_3_3_reduce')
inception_4a_3_3 = conv_2d(inception_4a_3_3_reduce, 208, filter_size=3, activation='relu', name='inception_4a_3_3')
inception_4a_5_5_reduce = conv_2d(pool3_3_3, 16, filter_size=1, activation='relu', name='inception_4a_5_5_reduce')
inception_4a_5_5 = conv_2d(inception_4a_5_5_reduce, 48, filter_size=5, activation='relu', name='inception_4a_5_5')
inception_4a_pool = max_pool_2d(pool3_3_3, kernel_size=3, strides=1, name='inception_4a_pool')
inception_4a_pool_1_1 = conv_2d(inception_4a_pool, 64, filter_size=1, activation='relu', name='inception_4a_pool_1_1')
inception_4a_output = merge([inception_4a_1_1, inception_4a_3_3, inception_4a_5_5, inception_4a_pool_1_1], mode='concat', axis=3, name='inception_4a_output')
# 4b
inception_4b_1_1 = conv_2d(inception_4a_output, 160, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4b_3_3_reduce = conv_2d(inception_4a_output, 112, filter_size=1, activation='relu', name='inception_4b_3_3_reduce')
inception_4b_3_3 = conv_2d(inception_4b_3_3_reduce, 224, filter_size=3, activation='relu', name='inception_4b_3_3')
inception_4b_5_5_reduce = conv_2d(inception_4a_output, 24, filter_size=1, activation='relu', name='inception_4b_5_5_reduce')
inception_4b_5_5 = conv_2d(inception_4b_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4b_5_5')
inception_4b_pool = max_pool_2d(inception_4a_output, kernel_size=3, strides=1, name='inception_4b_pool')
inception_4b_pool_1_1 = conv_2d(inception_4b_pool, 64, filter_size=1, activation='relu', name='inception_4b_pool_1_1')
inception_4b_output = merge([inception_4b_1_1, inception_4b_3_3, inception_4b_5_5, inception_4b_pool_1_1], mode='concat', axis=3, name='inception_4b_output')
# 4c
inception_4c_1_1 = conv_2d(inception_4b_output, 128, filter_size=1, activation='relu', name='inception_4c_1_1')
inception_4c_3_3_reduce = conv_2d(inception_4b_output, 128, filter_size=1, activation='relu', name='inception_4c_3_3_reduce')
inception_4c_3_3 = conv_2d(inception_4c_3_3_reduce, 256, filter_size=3, activation='relu', name='inception_4c_3_3')
inception_4c_5_5_reduce = conv_2d(inception_4b_output, 24, filter_size=1, activation='relu', name='inception_4c_5_5_reduce')
inception_4c_5_5 = conv_2d(inception_4c_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4c_5_5')
inception_4c_pool = max_pool_2d(inception_4b_output, kernel_size=3, strides=1)
inception_4c_pool_1_1 = conv_2d(inception_4c_pool, 64, filter_size=1, activation='relu', name='inception_4c_pool_1_1')
inception_4c_output = merge([inception_4c_1_1, inception_4c_3_3, inception_4c_5_5, inception_4c_pool_1_1], mode='concat', axis=3, name='inception_4c_output')
# 4d
inception_4d_1_1 = conv_2d(inception_4c_output, 112, filter_size=1, activation='relu', name='inception_4d_1_1')
inception_4d_3_3_reduce = conv_2d(inception_4c_output, 144, filter_size=1, activation='relu', name='inception_4d_3_3_reduce')
inception_4d_3_3 = conv_2d(inception_4d_3_3_reduce, 288, filter_size=3, activation='relu', name='inception_4d_3_3')
inception_4d_5_5_reduce = conv_2d(inception_4c_output, 32, filter_size=1, activation='relu', name='inception_4d_5_5_reduce')
inception_4d_5_5 = conv_2d(inception_4d_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4d_5_5')
inception_4d_pool = max_pool_2d(inception_4c_output, kernel_size=3, strides=1, name='inception_4d_pool')
inception_4d_pool_1_1 = conv_2d(inception_4d_pool, 64, filter_size=1, activation='relu', name='inception_4d_pool_1_1')
inception_4d_output = merge([inception_4d_1_1, inception_4d_3_3, inception_4d_5_5, inception_4d_pool_1_1], mode='concat', axis=3, name='inception_4d_output')
# 4e
inception_4e_1_1 = conv_2d(inception_4d_output, 256, filter_size=1, activation='relu', name='inception_4e_1_1')
inception_4e_3_3_reduce = conv_2d(inception_4d_output, 160, filter_size=1, activation='relu', name='inception_4e_3_3_reduce')
inception_4e_3_3 = conv_2d(inception_4e_3_3_reduce, 320, filter_size=3, activation='relu', name='inception_4e_3_3')
inception_4e_5_5_reduce = conv_2d(inception_4d_output, 32, filter_size=1, activation='relu', name='inception_4e_5_5_reduce')
inception_4e_5_5 = conv_2d(inception_4e_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_4e_5_5')
inception_4e_pool = max_pool_2d(inception_4d_output, kernel_size=3, strides=1, name='inception_4e_pool')
inception_4e_pool_1_1 = conv_2d(inception_4e_pool, 128, filter_size=1, activation='relu', name='inception_4e_pool_1_1')
inception_4e_output = merge([inception_4e_1_1, inception_4e_3_3, inception_4e_5_5, inception_4e_pool_1_1], axis=3, mode='concat')
pool4_3_3 = max_pool_2d(inception_4e_output, kernel_size=3, strides=2, name='pool_3_3')
# 5a
inception_5a_1_1 = conv_2d(pool4_3_3, 256, filter_size=1, activation='relu', name='inception_5a_1_1')
inception_5a_3_3_reduce = conv_2d(pool4_3_3, 160, filter_size=1, activation='relu', name='inception_5a_3_3_reduce')
inception_5a_3_3 = conv_2d(inception_5a_3_3_reduce, 320, filter_size=3, activation='relu', name='inception_5a_3_3')
inception_5a_5_5_reduce = conv_2d(pool4_3_3, 32, filter_size=1, activation='relu', name='inception_5a_5_5_reduce')
inception_5a_5_5 = conv_2d(inception_5a_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_5a_5_5')
inception_5a_pool = max_pool_2d(pool4_3_3, kernel_size=3, strides=1, name='inception_5a_pool')
inception_5a_pool_1_1 = conv_2d(inception_5a_pool, 128, filter_size=1, activation='relu', name='inception_5a_pool_1_1')
inception_5a_output = merge([inception_5a_1_1, inception_5a_3_3, inception_5a_5_5, inception_5a_pool_1_1], axis=3, mode='concat')
# 5b
inception_5b_1_1 = conv_2d(inception_5a_output, 384, filter_size=1, activation='relu', name='inception_5b_1_1')
inception_5b_3_3_reduce = conv_2d(inception_5a_output, 192, filter_size=1, activation='relu', name='inception_5b_3_3_reduce')
inception_5b_3_3 = conv_2d(inception_5b_3_3_reduce, 384, filter_size=3, activation='relu', name='inception_5b_3_3')
inception_5b_5_5_reduce = conv_2d(inception_5a_output, 48, filter_size=1, activation='relu', name='inception_5b_5_5_reduce')
inception_5b_5_5 = conv_2d(inception_5b_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_5b_5_5')
inception_5b_pool = max_pool_2d(inception_5a_output, kernel_size=3, strides=1, name='inception_5b_pool')
inception_5b_pool_1_1 = conv_2d(inception_5b_pool, 128, filter_size=1, activation='relu', name='inception_5b_pool_1_1')
inception_5b_output = merge([inception_5b_1_1, inception_5b_3_3, inception_5b_5_5, inception_5b_pool_1_1], axis=3, mode='concat')
pool5_7_7 = avg_pool_2d(inception_5b_output, kernel_size=7, strides=1)
pool5_7_7 = dropout(pool5_7_7, 0.4)
# fc
loss = fully_connected(pool5_7_7, len(yTest[0]), activation='softmax')
network = regression(loss, optimizer='momentum',
loss='categorical_crossentropy',
learning_rate=0.001)
# to train
model = tflearn.DNN(network, checkpoint_path='model_googlenet',
max_checkpoints=1, tensorboard_verbose=2)
model.load(_path)
pred = model.predict(xTest)
df = pd.DataFrame(pred)
df.to_csv(_path + ".csv")
newList = pred.copy()
newList = convert2(newList)
if _CSV: makeCSV(newList)
pred = convert2(pred)
pred = convert3(pred)
yTest = convert3(yTest)
print(metrics.confusion_matrix(yTest, pred))
print(metrics.classification_report(yTest, pred))
print('Accuracy', accuracy_score(yTest, pred))
print()
if _wrFile: writeTest(pred)
# pickle.dump((trainX,trainY,testX,testY,vocab_size),f)
#else:
# print("traning data exists in cache. going to use it.")
# 3.Building convolutional network
######################################MODEL:1.conv-2.conv-3.conv-4.max_pool-5.dropout-6.FC##############################################################################################
#(shape=None, placeholder=None, dtype=tf.float32,data_preprocessing=None, data_augmentation=None,name="InputData")
network = input_data(shape=[None, 100], name='input') #[None, 100] `input_data` is used as a data entry (placeholder) of a network. This placeholder will be feeded with data when training
network = tflearn.embedding(network, input_dim=vocab_size, output_dim=128) #TODO [None, 100,128].embedding layer for a sequence of ids. network: Incoming 2-D Tensor. input_dim: vocabulary size, oput_dim:embedding size
#conv_1d(incoming,nb_filter,filter_size)
branch1 = conv_1d(network, 128, 1, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 2, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2") # [batch_size, new steps1, nb_filters]. padding:"VALID",only ever drops the right-most columns
branch4 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2") # [batch_size, new steps2, nb_filters]
branch5 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2") # [batch_size, new steps3, nb_filters]
network = merge([branch1, branch2, branch3,branch4,branch5], mode='concat', axis=1) # merge a list of `Tensor` into a single one.===>[batch_size, new steps1+new step2+new step3, nb_filters]
network = tf.expand_dims(network, 2) #[batch_size, new steps1+new step2+new step3,1, nb_filters] Inserts a dimension of 1 into a tensor's shape
network = global_max_pool(network) #[batch_size, pooled dim]
network = dropout(network, 0.5) #[batch_size, pooled dim]
network = fully_connected(network, number_classes, activation='softmax') #matmul([batch_size, pooled_dim],[pooled_dim,2])---->[batch_size,number_classes]
top5 = tflearn.metrics.Top_k(k=5)
network = regression(network, optimizer='adam', learning_rate=0.001,loss='categorical_crossentropy', name='target') #metric=top5
######################################MODEL:1.conv-2.conv-3.conv-4.max_pool-5.dropout-6.FC################################################################################################
# 4.Training
model = tflearn.DNN(network, tensorboard_verbose=0)
#model.fit(trainX, trainY, n_epoch = 10, shuffle=True, validation_set=(testX, testY), show_metric=True, batch_size=256) #32
#model.save('model_zhihu_cnn12345')
model.load('model_zhihu_cnn12345')
print("going to make a prediction...")
predict_result=model.predict(testX[0:1000])
print("predict_result:",predict_result)
def _model4():
# Taken from TFLearn examples and based on Googles Inception. DO NOT RUN!!!!
global yTest, img_aug
tf.reset_default_graph()
img_prep = ImagePreprocessing()
img_prep.add_featurewise_zero_center()
img_prep.add_featurewise_stdnorm()
network = input_data(shape=[None, inputSize, inputSize, dim],
name='input',
data_preprocessing=img_prep,
data_augmentation=img_aug)
conv1_7_7 = conv_2d(network, 64, 7, strides=2, activation='relu', name='conv1_7_7_s2')
pool1_3_3 = max_pool_2d(conv1_7_7, 3, strides=2)
pool1_3_3 = local_response_normalization(pool1_3_3)
conv2_3_3_reduce = conv_2d(pool1_3_3, 64, 1, activation='relu', name='conv2_3_3_reduce')
conv2_3_3 = conv_2d(conv2_3_3_reduce, 192, 3, activation='relu', name='conv2_3_3')
conv2_3_3 = local_response_normalization(conv2_3_3)
pool2_3_3 = max_pool_2d(conv2_3_3, kernel_size=3, strides=2, name='pool2_3_3_s2')
# 3a
inception_3a_1_1 = conv_2d(pool2_3_3, 64, 1, activation='relu', name='inception_3a_1_1')
inception_3a_3_3_reduce = conv_2d(pool2_3_3, 96, 1, activation='relu', name='inception_3a_3_3_reduce')
inception_3a_3_3 = conv_2d(inception_3a_3_3_reduce, 128, filter_size=3, activation='relu', name='inception_3a_3_3')
inception_3a_5_5_reduce = conv_2d(pool2_3_3, 16, filter_size=1, activation='relu', name='inception_3a_5_5_reduce')
inception_3a_5_5 = conv_2d(inception_3a_5_5_reduce, 32, filter_size=5, activation='relu', name='inception_3a_5_5')
inception_3a_pool = max_pool_2d(pool2_3_3, kernel_size=3, strides=1, name='inception_3a_pool')
inception_3a_pool_1_1 = conv_2d(inception_3a_pool, 32, filter_size=1, activation='relu', name='inception_3a_pool_1_1')
inception_3a_output = merge([inception_3a_1_1, inception_3a_3_3, inception_3a_5_5, inception_3a_pool_1_1], mode='concat', axis=3)
# 3b
inception_3b_1_1 = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_1_1')
inception_3b_3_3_reduce = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_3_3_reduce')
inception_3b_3_3 = conv_2d(inception_3b_3_3_reduce, 192, filter_size=3, activation='relu', name='inception_3b_3_3')
inception_3b_5_5_reduce = conv_2d(inception_3a_output, 32, filter_size=1, activation='relu', name='inception_3b_5_5_reduce')
inception_3b_5_5 = conv_2d(inception_3b_5_5_reduce, 96, filter_size=5, name='inception_3b_5_5')
inception_3b_pool = max_pool_2d(inception_3a_output, kernel_size=3, strides=1, name='inception_3b_pool')
inception_3b_pool_1_1 = conv_2d(inception_3b_pool, 64, filter_size=1, activation='relu', name='inception_3b_pool_1_1')
inception_3b_output = merge([inception_3b_1_1, inception_3b_3_3, inception_3b_5_5, inception_3b_pool_1_1], mode='concat', axis=3, name='inception_3b_output')
pool3_3_3 = max_pool_2d(inception_3b_output, kernel_size=3, strides=2, name='pool3_3_3')
# 4a
inception_4a_1_1 = conv_2d(pool3_3_3, 192, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4a_3_3_reduce = conv_2d(pool3_3_3, 96, filter_size=1, activation='relu', name='inception_4a_3_3_reduce')
inception_4a_3_3 = conv_2d(inception_4a_3_3_reduce, 208, filter_size=3, activation='relu', name='inception_4a_3_3')
inception_4a_5_5_reduce = conv_2d(pool3_3_3, 16, filter_size=1, activation='relu', name='inception_4a_5_5_reduce')
inception_4a_5_5 = conv_2d(inception_4a_5_5_reduce, 48, filter_size=5, activation='relu', name='inception_4a_5_5')
inception_4a_pool = max_pool_2d(pool3_3_3, kernel_size=3, strides=1, name='inception_4a_pool')
inception_4a_pool_1_1 = conv_2d(inception_4a_pool, 64, filter_size=1, activation='relu', name='inception_4a_pool_1_1')
inception_4a_output = merge([inception_4a_1_1, inception_4a_3_3, inception_4a_5_5, inception_4a_pool_1_1], mode='concat', axis=3, name='inception_4a_output')
# 4b
inception_4b_1_1 = conv_2d(inception_4a_output, 160, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4b_3_3_reduce = conv_2d(inception_4a_output, 112, filter_size=1, activation='relu', name='inception_4b_3_3_reduce')
inception_4b_3_3 = conv_2d(inception_4b_3_3_reduce, 224, filter_size=3, activation='relu', name='inception_4b_3_3')
inception_4b_5_5_reduce = conv_2d(inception_4a_output, 24, filter_size=1, activation='relu', name='inception_4b_5_5_reduce')
inception_4b_5_5 = conv_2d(inception_4b_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4b_5_5')
inception_4b_pool = max_pool_2d(inception_4a_output, kernel_size=3, strides=1, name='inception_4b_pool')
inception_4b_pool_1_1 = conv_2d(inception_4b_pool, 64, filter_size=1, activation='relu', name='inception_4b_pool_1_1')
inception_4b_output = merge([inception_4b_1_1, inception_4b_3_3, inception_4b_5_5, inception_4b_pool_1_1], mode='concat', axis=3, name='inception_4b_output')
# 4c
inception_4c_1_1 = conv_2d(inception_4b_output, 128, filter_size=1, activation='relu', name='inception_4c_1_1')
inception_4c_3_3_reduce = conv_2d(inception_4b_output, 128, filter_size=1, activation='relu', name='inception_4c_3_3_reduce')
inception_4c_3_3 = conv_2d(inception_4c_3_3_reduce, 256, filter_size=3, activation='relu', name='inception_4c_3_3')
inception_4c_5_5_reduce = conv_2d(inception_4b_output, 24, filter_size=1, activation='relu', name='inception_4c_5_5_reduce')
inception_4c_5_5 = conv_2d(inception_4c_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4c_5_5')
inception_4c_pool = max_pool_2d(inception_4b_output, kernel_size=3, strides=1)
inception_4c_pool_1_1 = conv_2d(inception_4c_pool, 64, filter_size=1, activation='relu', name='inception_4c_pool_1_1')
inception_4c_output = merge([inception_4c_1_1, inception_4c_3_3, inception_4c_5_5, inception_4c_pool_1_1], mode='concat', axis=3, name='inception_4c_output')
# 4d
inception_4d_1_1 = conv_2d(inception_4c_output, 112, filter_size=1, activation='relu', name='inception_4d_1_1')
inception_4d_3_3_reduce = conv_2d(inception_4c_output, 144, filter_size=1, activation='relu', name='inception_4d_3_3_reduce')
inception_4d_3_3 = conv_2d(inception_4d_3_3_reduce, 288, filter_size=3, activation='relu', name='inception_4d_3_3')
inception_4d_5_5_reduce = conv_2d(inception_4c_output, 32, filter_size=1, activation='relu', name='inception_4d_5_5_reduce')
inception_4d_5_5 = conv_2d(inception_4d_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4d_5_5')
inception_4d_pool = max_pool_2d(inception_4c_output, kernel_size=3, strides=1, name='inception_4d_pool')
inception_4d_pool_1_1 = conv_2d(inception_4d_pool, 64, filter_size=1, activation='relu', name='inception_4d_pool_1_1')
inception_4d_output = merge([inception_4d_1_1, inception_4d_3_3, inception_4d_5_5, inception_4d_pool_1_1], mode='concat', axis=3, name='inception_4d_output')
# 4e
inception_4e_1_1 = conv_2d(inception_4d_output, 256, filter_size=1, activation='relu', name='inception_4e_1_1')
inception_4e_3_3_reduce = conv_2d(inception_4d_output, 160, filter_size=1, activation='relu', name='inception_4e_3_3_reduce')
inception_4e_3_3 = conv_2d(inception_4e_3_3_reduce, 320, filter_size=3, activation='relu', name='inception_4e_3_3')
inception_4e_5_5_reduce = conv_2d(inception_4d_output, 32, filter_size=1, activation='relu', name='inception_4e_5_5_reduce')
inception_4e_5_5 = conv_2d(inception_4e_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_4e_5_5')
inception_4e_pool = max_pool_2d(inception_4d_output, kernel_size=3, strides=1, name='inception_4e_pool')
inception_4e_pool_1_1 = conv_2d(inception_4e_pool, 128, filter_size=1, activation='relu', name='inception_4e_pool_1_1')
inception_4e_output = merge([inception_4e_1_1, inception_4e_3_3, inception_4e_5_5, inception_4e_pool_1_1], axis=3, mode='concat')
pool4_3_3 = max_pool_2d(inception_4e_output, kernel_size=3, strides=2, name='pool_3_3')
# 5a
inception_5a_1_1 = conv_2d(pool4_3_3, 256, filter_size=1, activation='relu', name='inception_5a_1_1')
inception_5a_3_3_reduce = conv_2d(pool4_3_3, 160, filter_size=1, activation='relu', name='inception_5a_3_3_reduce')
inception_5a_3_3 = conv_2d(inception_5a_3_3_reduce, 320, filter_size=3, activation='relu', name='inception_5a_3_3')
inception_5a_5_5_reduce = conv_2d(pool4_3_3, 32, filter_size=1, activation='relu', name='inception_5a_5_5_reduce')
inception_5a_5_5 = conv_2d(inception_5a_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_5a_5_5')
inception_5a_pool = max_pool_2d(pool4_3_3, kernel_size=3, strides=1, name='inception_5a_pool')
inception_5a_pool_1_1 = conv_2d(inception_5a_pool, 128, filter_size=1, activation='relu', name='inception_5a_pool_1_1')
inception_5a_output = merge([inception_5a_1_1, inception_5a_3_3, inception_5a_5_5, inception_5a_pool_1_1], axis=3, mode='concat')
# 5b
inception_5b_1_1 = conv_2d(inception_5a_output, 384, filter_size=1, activation='relu', name='inception_5b_1_1')
inception_5b_3_3_reduce = conv_2d(inception_5a_output, 192, filter_size=1, activation='relu', name='inception_5b_3_3_reduce')
inception_5b_3_3 = conv_2d(inception_5b_3_3_reduce, 384, filter_size=3, activation='relu', name='inception_5b_3_3')
inception_5b_5_5_reduce = conv_2d(inception_5a_output, 48, filter_size=1, activation='relu', name='inception_5b_5_5_reduce')
inception_5b_5_5 = conv_2d(inception_5b_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_5b_5_5')
inception_5b_pool = max_pool_2d(inception_5a_output, kernel_size=3, strides=1, name='inception_5b_pool')
inception_5b_pool_1_1 = conv_2d(inception_5b_pool, 128, filter_size=1, activation='relu', name='inception_5b_pool_1_1')
inception_5b_output = merge([inception_5b_1_1, inception_5b_3_3, inception_5b_5_5, inception_5b_pool_1_1], axis=3, mode='concat')
pool5_7_7 = avg_pool_2d(inception_5b_output, kernel_size=7, strides=1)
pool5_7_7 = dropout(pool5_7_7, 0.4)
# fc
loss = fully_connected(pool5_7_7, len(Y[0]), activation='softmax')
network = regression(loss, optimizer='momentum',
loss='categorical_crossentropy',
learning_rate=0.001)
# to train
model = tflearn.DNN(network, checkpoint_path='model_googlenet',
max_checkpoints=1, tensorboard_verbose=2)
model.fit(X, Y, n_epoch=epochNum, validation_set=(xTest, yTest), shuffle=True,
show_metric=True, batch_size=batchNum, snapshot_step=200,
snapshot_epoch=False, run_id='googlenet_oxflowers17')
if modelStore: model.save(_id + '-model.tflearn')
def build_le_network(self, network):
""" Build google net. (google inception v3)
Args:
network:
Returns:
"""
from tflearn.layers.merge_ops import merge
conv1_7_7 = conv_2d(network, 64, 7, strides=2, activation='relu',
name='conv1_7_7_s2')
pool1_3_3 = max_pool_2d(conv1_7_7, 3, strides=2)
pool1_3_3 = local_response_normalization(pool1_3_3)
conv2_3_3_reduce = conv_2d(pool1_3_3, 64, 1, activation='relu',
name='conv2_3_3_reduce')
conv2_3_3 = conv_2d(conv2_3_3_reduce, 192, 3, activation='relu',
name='conv2_3_3')
conv2_3_3 = local_response_normalization(conv2_3_3)
pool2_3_3 = max_pool_2d(conv2_3_3, kernel_size=3, strides=2,
name='pool2_3_3_s2')
inception_3a_1_1 = conv_2d(pool2_3_3, 64, 1, activation='relu',
name='inception_3a_1_1')
inception_3a_3_3_reduce = conv_2d(pool2_3_3, 96, 1, activation='relu',
name='inception_3a_3_3_reduce')
inception_3a_3_3 = conv_2d(inception_3a_3_3_reduce, 128, filter_size=3,
activation='relu', name='inception_3a_3_3')
inception_3a_5_5_reduce = conv_2d(pool2_3_3, 16, filter_size=1,
activation='relu',
name='inception_3a_5_5_reduce')
inception_3a_5_5 = conv_2d(inception_3a_5_5_reduce, 32, filter_size=5,
activation='relu', name='inception_3a_5_5')
inception_3a_pool = max_pool_2d(pool2_3_3, kernel_size=3, strides=1, )
inception_3a_pool_1_1 = conv_2d(inception_3a_pool, 32, filter_size=1,
activation='relu',
name='inception_3a_pool_1_1')
# merge the inception_3a__
inception_3a_output = merge(
[inception_3a_1_1, inception_3a_3_3, inception_3a_5_5,
inception_3a_pool_1_1], mode='concat', axis=3)
inception_3b_1_1 = conv_2d(inception_3a_output, 128, filter_size=1,
activation='relu', name='inception_3b_1_1')
inception_3b_3_3_reduce = conv_2d(inception_3a_output, 128,
filter_size=1, activation='relu',
name='inception_3b_3_3_reduce')
inception_3b_3_3 = conv_2d(inception_3b_3_3_reduce, 192, filter_size=3,
activation='relu', name='inception_3b_3_3')
inception_3b_5_5_reduce = conv_2d(inception_3a_output, 32,
filter_size=1, activation='relu',
name='inception_3b_5_5_reduce')
inception_3b_5_5 = conv_2d(inception_3b_5_5_reduce, 96, filter_size=5,
name='inception_3b_5_5')
inception_3b_pool = max_pool_2d(inception_3a_output, kernel_size=3,
strides=1, name='inception_3b_pool')
inception_3b_pool_1_1 = conv_2d(inception_3b_pool, 64, filter_size=1,
activation='relu',
name='inception_3b_pool_1_1')
# merge the inception_3b_*
inception_3b_output = merge(
[inception_3b_1_1, inception_3b_3_3, inception_3b_5_5,
inception_3b_pool_1_1], mode='concat', axis=3,
name='inception_3b_output')
pool3_3_3 = max_pool_2d(inception_3b_output, kernel_size=3, strides=2,
name='pool3_3_3')
inception_4a_1_1 = conv_2d(pool3_3_3, 192, filter_size=1,
activation='relu', name='inception_4a_1_1')
inception_4a_3_3_reduce = conv_2d(pool3_3_3, 96, filter_size=1,
activation='relu',
name='inception_4a_3_3_reduce')
inception_4a_3_3 = conv_2d(inception_4a_3_3_reduce, 208, filter_size=3,
activation='relu', name='inception_4a_3_3')
inception_4a_5_5_reduce = conv_2d(pool3_3_3, 16, filter_size=1,
activation='relu',
name='inception_4a_5_5_reduce')
inception_4a_5_5 = conv_2d(inception_4a_5_5_reduce, 48, filter_size=5,
activation='relu', name='inception_4a_5_5')
inception_4a_pool = max_pool_2d(pool3_3_3, kernel_size=3, strides=1,
name='inception_4a_pool')
inception_4a_pool_1_1 = conv_2d(inception_4a_pool, 64, filter_size=1,
activation='relu',
name='inception_4a_pool_1_1')
inception_4a_output = merge(
[inception_4a_1_1, inception_4a_3_3, inception_4a_5_5,
inception_4a_pool_1_1], mode='concat', axis=3,
name='inception_4a_output')
inception_4b_1_1 = conv_2d(inception_4a_output, 160, filter_size=1,
activation='relu', name='inception_4a_1_1')
inception_4b_3_3_reduce = conv_2d(inception_4a_output, 112,
filter_size=1, activation='relu',
name='inception_4b_3_3_reduce')
inception_4b_3_3 = conv_2d(inception_4b_3_3_reduce, 224, filter_size=3,
activation='relu', name='inception_4b_3_3')
inception_4b_5_5_reduce = conv_2d(inception_4a_output, 24,
filter_size=1, activation='relu',
name='inception_4b_5_5_reduce')
inception_4b_5_5 = conv_2d(inception_4b_5_5_reduce, 64, filter_size=5,
activation='relu', name='inception_4b_5_5')
inception_4b_pool = max_pool_2d(inception_4a_output, kernel_size=3,
strides=1, name='inception_4b_pool')
inception_4b_pool_1_1 = conv_2d(inception_4b_pool, 64, filter_size=1,
activation='relu',
name='inception_4b_pool_1_1')
inception_4b_output = merge(
[inception_4b_1_1, inception_4b_3_3, inception_4b_5_5,
inception_4b_pool_1_1], mode='concat', axis=3,
name='inception_4b_output')
inception_4c_1_1 = conv_2d(inception_4b_output, 128, filter_size=1,
activation='relu', name='inception_4c_1_1')
inception_4c_3_3_reduce = conv_2d(inception_4b_output, 128,
filter_size=1, activation='relu',
name='inception_4c_3_3_reduce')
inception_4c_3_3 = conv_2d(inception_4c_3_3_reduce, 256, filter_size=3,
activation='relu', name='inception_4c_3_3')
inception_4c_5_5_reduce = conv_2d(inception_4b_output, 24,
filter_size=1, activation='relu',
name='inception_4c_5_5_reduce')
inception_4c_5_5 = conv_2d(inception_4c_5_5_reduce, 64, filter_size=5,
activation='relu', name='inception_4c_5_5')
inception_4c_pool = max_pool_2d(inception_4b_output, kernel_size=3,
strides=1)
inception_4c_pool_1_1 = conv_2d(inception_4c_pool, 64, filter_size=1,
activation='relu',
name='inception_4c_pool_1_1')
inception_4c_output = merge(
[inception_4c_1_1, inception_4c_3_3, inception_4c_5_5,
inception_4c_pool_1_1], mode='concat', axis=3,
name='inception_4c_output')
inception_4d_1_1 = conv_2d(inception_4c_output, 112, filter_size=1,
activation='relu', name='inception_4d_1_1')
inception_4d_3_3_reduce = conv_2d(inception_4c_output, 144,
filter_size=1, activation='relu',
name='inception_4d_3_3_reduce')
inception_4d_3_3 = conv_2d(inception_4d_3_3_reduce, 288, filter_size=3,
activation='relu', name='inception_4d_3_3')
inception_4d_5_5_reduce = conv_2d(inception_4c_output, 32,
filter_size=1, activation='relu',
name='inception_4d_5_5_reduce')
inception_4d_5_5 = conv_2d(inception_4d_5_5_reduce, 64, filter_size=5,
activation='relu', name='inception_4d_5_5')
inception_4d_pool = max_pool_2d(inception_4c_output, kernel_size=3,
strides=1, name='inception_4d_pool')
inception_4d_pool_1_1 = conv_2d(inception_4d_pool, 64, filter_size=1,
activation='relu',
name='inception_4d_pool_1_1')
inception_4d_output = merge(
[inception_4d_1_1, inception_4d_3_3, inception_4d_5_5,
inception_4d_pool_1_1], mode='concat', axis=3,
name='inception_4d_output')
inception_4e_1_1 = conv_2d(inception_4d_output, 256, filter_size=1,
activation='relu', name='inception_4e_1_1')
inception_4e_3_3_reduce = conv_2d(inception_4d_output, 160,
filter_size=1, activation='relu',
name='inception_4e_3_3_reduce')
inception_4e_3_3 = conv_2d(inception_4e_3_3_reduce, 320, filter_size=3,
activation='relu', name='inception_4e_3_3')
inception_4e_5_5_reduce = conv_2d(inception_4d_output, 32,
filter_size=1, activation='relu',
name='inception_4e_5_5_reduce')
inception_4e_5_5 = conv_2d(inception_4e_5_5_reduce, 128, filter_size=5,
activation='relu', name='inception_4e_5_5')
inception_4e_pool = max_pool_2d(inception_4d_output, kernel_size=3,
strides=1, name='inception_4e_pool')
inception_4e_pool_1_1 = conv_2d(inception_4e_pool, 128, filter_size=1,
activation='relu',
name='inception_4e_pool_1_1')
inception_4e_output = merge(
[inception_4e_1_1, inception_4e_3_3, inception_4e_5_5,
inception_4e_pool_1_1], axis=3, mode='concat')
pool4_3_3 = max_pool_2d(inception_4e_output, kernel_size=3, strides=2,
name='pool_3_3')
inception_5a_1_1 = conv_2d(pool4_3_3, 256, filter_size=1,
activation='relu', name='inception_5a_1_1')
inception_5a_3_3_reduce = conv_2d(pool4_3_3, 160, filter_size=1,
activation='relu',
name='inception_5a_3_3_reduce')
inception_5a_3_3 = conv_2d(inception_5a_3_3_reduce, 320, filter_size=3,
activation='relu', name='inception_5a_3_3')
inception_5a_5_5_reduce = conv_2d(pool4_3_3, 32, filter_size=1,
activation='relu',
name='inception_5a_5_5_reduce')
inception_5a_5_5 = conv_2d(inception_5a_5_5_reduce, 128, filter_size=5,
activation='relu', name='inception_5a_5_5')
inception_5a_pool = max_pool_2d(pool4_3_3, kernel_size=3, strides=1,
name='inception_5a_pool')
inception_5a_pool_1_1 = conv_2d(inception_5a_pool, 128, filter_size=1,
activation='relu',
name='inception_5a_pool_1_1')
inception_5a_output = merge(
[inception_5a_1_1, inception_5a_3_3, inception_5a_5_5,
inception_5a_pool_1_1], axis=3, mode='concat')
inception_5b_1_1 = conv_2d(inception_5a_output, 384, filter_size=1,
activation='relu', name='inception_5b_1_1')
inception_5b_3_3_reduce = conv_2d(inception_5a_output, 192,
filter_size=1, activation='relu',
name='inception_5b_3_3_reduce')
inception_5b_3_3 = conv_2d(inception_5b_3_3_reduce, 384, filter_size=3,
activation='relu', name='inception_5b_3_3')
inception_5b_5_5_reduce = conv_2d(inception_5a_output, 48,
filter_size=1, activation='relu',
name='inception_5b_5_5_reduce')
inception_5b_5_5 = conv_2d(inception_5b_5_5_reduce, 128, filter_size=5,
activation='relu', name='inception_5b_5_5')
inception_5b_pool = max_pool_2d(inception_5a_output, kernel_size=3,
strides=1, name='inception_5b_pool')
inception_5b_pool_1_1 = conv_2d(inception_5b_pool, 128, filter_size=1,
activation='relu',
name='inception_5b_pool_1_1')
inception_5b_output = merge(
[inception_5b_1_1, inception_5b_3_3, inception_5b_5_5,
inception_5b_pool_1_1], axis=3, mode='concat')
pool5_7_7 = avg_pool_2d(inception_5b_output, kernel_size=7, strides=1)
pool5_7_7 = dropout(pool5_7_7, 0.4)
loss = fully_connected(pool5_7_7, 2, activation='softmax')
network = regression(loss, optimizer='momentum',
loss='categorical_crossentropy',
learning_rate=0.001)
return network
def inception_v3(width, height, frame_count, lr, output=9, model_name = 'sentnet_color.model'):
network = input_data(shape=[None, width, height,3], name='input')
conv1_7_7 = conv_2d(network, 64, 7, strides=2, activation='relu', name = 'conv1_7_7_s2')
pool1_3_3 = max_pool_2d(conv1_7_7, 3,strides=2)
pool1_3_3 = local_response_normalization(pool1_3_3)
conv2_3_3_reduce = conv_2d(pool1_3_3, 64,1, activation='relu',name = 'conv2_3_3_reduce')
conv2_3_3 = conv_2d(conv2_3_3_reduce, 192,3, activation='relu', name='conv2_3_3')
conv2_3_3 = local_response_normalization(conv2_3_3)
pool2_3_3 = max_pool_2d(conv2_3_3, kernel_size=3, strides=2, name='pool2_3_3_s2')
inception_3a_1_1 = conv_2d(pool2_3_3, 64, 1, activation='relu', name='inception_3a_1_1')
inception_3a_3_3_reduce = conv_2d(pool2_3_3, 96,1, activation='relu', name='inception_3a_3_3_reduce')
inception_3a_3_3 = conv_2d(inception_3a_3_3_reduce, 128,filter_size=3, activation='relu', name = 'inception_3a_3_3')
inception_3a_5_5_reduce = conv_2d(pool2_3_3,16, filter_size=1,activation='relu', name ='inception_3a_5_5_reduce' )
inception_3a_5_5 = conv_2d(inception_3a_5_5_reduce, 32, filter_size=5, activation='relu', name= 'inception_3a_5_5')
inception_3a_pool = max_pool_2d(pool2_3_3, kernel_size=3, strides=1, )
inception_3a_pool_1_1 = conv_2d(inception_3a_pool, 32, filter_size=1, activation='relu', name='inception_3a_pool_1_1')
# merge the inception_3a__
inception_3a_output = merge([inception_3a_1_1, inception_3a_3_3, inception_3a_5_5, inception_3a_pool_1_1], mode='concat', axis=3)
inception_3b_1_1 = conv_2d(inception_3a_output, 128,filter_size=1,activation='relu', name= 'inception_3b_1_1' )
inception_3b_3_3_reduce = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_3_3_reduce')
inception_3b_3_3 = conv_2d(inception_3b_3_3_reduce, 192, filter_size=3, activation='relu',name='inception_3b_3_3')
inception_3b_5_5_reduce = conv_2d(inception_3a_output, 32, filter_size=1, activation='relu', name = 'inception_3b_5_5_reduce')
inception_3b_5_5 = conv_2d(inception_3b_5_5_reduce, 96, filter_size=5, name = 'inception_3b_5_5')
inception_3b_pool = max_pool_2d(inception_3a_output, kernel_size=3, strides=1, name='inception_3b_pool')
inception_3b_pool_1_1 = conv_2d(inception_3b_pool, 64, filter_size=1,activation='relu', name='inception_3b_pool_1_1')
#merge the inception_3b_*
inception_3b_output = merge([inception_3b_1_1, inception_3b_3_3, inception_3b_5_5, inception_3b_pool_1_1], mode='concat',axis=3,name='inception_3b_output')
pool3_3_3 = max_pool_2d(inception_3b_output, kernel_size=3, strides=2, name='pool3_3_3')
inception_4a_1_1 = conv_2d(pool3_3_3, 192, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4a_3_3_reduce = conv_2d(pool3_3_3, 96, filter_size=1, activation='relu', name='inception_4a_3_3_reduce')
inception_4a_3_3 = conv_2d(inception_4a_3_3_reduce, 208, filter_size=3, activation='relu', name='inception_4a_3_3')
inception_4a_5_5_reduce = conv_2d(pool3_3_3, 16, filter_size=1, activation='relu', name='inception_4a_5_5_reduce')
inception_4a_5_5 = conv_2d(inception_4a_5_5_reduce, 48, filter_size=5, activation='relu', name='inception_4a_5_5')
inception_4a_pool = max_pool_2d(pool3_3_3, kernel_size=3, strides=1, name='inception_4a_pool')
inception_4a_pool_1_1 = conv_2d(inception_4a_pool, 64, filter_size=1, activation='relu', name='inception_4a_pool_1_1')
inception_4a_output = merge([inception_4a_1_1, inception_4a_3_3, inception_4a_5_5, inception_4a_pool_1_1], mode='concat', axis=3, name='inception_4a_output')
inception_4b_1_1 = conv_2d(inception_4a_output, 160, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4b_3_3_reduce = conv_2d(inception_4a_output, 112, filter_size=1, activation='relu', name='inception_4b_3_3_reduce')
inception_4b_3_3 = conv_2d(inception_4b_3_3_reduce, 224, filter_size=3, activation='relu', name='inception_4b_3_3')
inception_4b_5_5_reduce = conv_2d(inception_4a_output, 24, filter_size=1, activation='relu', name='inception_4b_5_5_reduce')
inception_4b_5_5 = conv_2d(inception_4b_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4b_5_5')
inception_4b_pool = max_pool_2d(inception_4a_output, kernel_size=3, strides=1, name='inception_4b_pool')
inception_4b_pool_1_1 = conv_2d(inception_4b_pool, 64, filter_size=1, activation='relu', name='inception_4b_pool_1_1')
inception_4b_output = merge([inception_4b_1_1, inception_4b_3_3, inception_4b_5_5, inception_4b_pool_1_1], mode='concat', axis=3, name='inception_4b_output')
inception_4c_1_1 = conv_2d(inception_4b_output, 128, filter_size=1, activation='relu',name='inception_4c_1_1')
inception_4c_3_3_reduce = conv_2d(inception_4b_output, 128, filter_size=1, activation='relu', name='inception_4c_3_3_reduce')
inception_4c_3_3 = conv_2d(inception_4c_3_3_reduce, 256, filter_size=3, activation='relu', name='inception_4c_3_3')
inception_4c_5_5_reduce = conv_2d(inception_4b_output, 24, filter_size=1, activation='relu', name='inception_4c_5_5_reduce')
inception_4c_5_5 = conv_2d(inception_4c_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4c_5_5')
inception_4c_pool = max_pool_2d(inception_4b_output, kernel_size=3, strides=1)
inception_4c_pool_1_1 = conv_2d(inception_4c_pool, 64, filter_size=1, activation='relu', name='inception_4c_pool_1_1')
inception_4c_output = merge([inception_4c_1_1, inception_4c_3_3, inception_4c_5_5, inception_4c_pool_1_1], mode='concat', axis=3,name='inception_4c_output')
inception_4d_1_1 = conv_2d(inception_4c_output, 112, filter_size=1, activation='relu', name='inception_4d_1_1')
inception_4d_3_3_reduce = conv_2d(inception_4c_output, 144, filter_size=1, activation='relu', name='inception_4d_3_3_reduce')
inception_4d_3_3 = conv_2d(inception_4d_3_3_reduce, 288, filter_size=3, activation='relu', name='inception_4d_3_3')
inception_4d_5_5_reduce = conv_2d(inception_4c_output, 32, filter_size=1, activation='relu', name='inception_4d_5_5_reduce')
inception_4d_5_5 = conv_2d(inception_4d_5_5_reduce, 64, filter_size=5, activation='relu', name='inception_4d_5_5')
inception_4d_pool = max_pool_2d(inception_4c_output, kernel_size=3, strides=1, name='inception_4d_pool')
inception_4d_pool_1_1 = conv_2d(inception_4d_pool, 64, filter_size=1, activation='relu', name='inception_4d_pool_1_1')
inception_4d_output = merge([inception_4d_1_1, inception_4d_3_3, inception_4d_5_5, inception_4d_pool_1_1], mode='concat', axis=3, name='inception_4d_output')
inception_4e_1_1 = conv_2d(inception_4d_output, 256, filter_size=1, activation='relu', name='inception_4e_1_1')
inception_4e_3_3_reduce = conv_2d(inception_4d_output, 160, filter_size=1, activation='relu', name='inception_4e_3_3_reduce')
inception_4e_3_3 = conv_2d(inception_4e_3_3_reduce, 320, filter_size=3, activation='relu', name='inception_4e_3_3')
inception_4e_5_5_reduce = conv_2d(inception_4d_output, 32, filter_size=1, activation='relu', name='inception_4e_5_5_reduce')
inception_4e_5_5 = conv_2d(inception_4e_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_4e_5_5')
inception_4e_pool = max_pool_2d(inception_4d_output, kernel_size=3, strides=1, name='inception_4e_pool')
inception_4e_pool_1_1 = conv_2d(inception_4e_pool, 128, filter_size=1, activation='relu', name='inception_4e_pool_1_1')
inception_4e_output = merge([inception_4e_1_1, inception_4e_3_3, inception_4e_5_5,inception_4e_pool_1_1],axis=3, mode='concat')
pool4_3_3 = max_pool_2d(inception_4e_output, kernel_size=3, strides=2, name='pool_3_3')
inception_5a_1_1 = conv_2d(pool4_3_3, 256, filter_size=1, activation='relu', name='inception_5a_1_1')
inception_5a_3_3_reduce = conv_2d(pool4_3_3, 160, filter_size=1, activation='relu', name='inception_5a_3_3_reduce')
inception_5a_3_3 = conv_2d(inception_5a_3_3_reduce, 320, filter_size=3, activation='relu', name='inception_5a_3_3')
inception_5a_5_5_reduce = conv_2d(pool4_3_3, 32, filter_size=1, activation='relu', name='inception_5a_5_5_reduce')
inception_5a_5_5 = conv_2d(inception_5a_5_5_reduce, 128, filter_size=5, activation='relu', name='inception_5a_5_5')
inception_5a_pool = max_pool_2d(pool4_3_3, kernel_size=3, strides=1, name='inception_5a_pool')
inception_5a_pool_1_1 = conv_2d(inception_5a_pool, 128, filter_size=1,activation='relu', name='inception_5a_pool_1_1')
inception_5a_output = merge([inception_5a_1_1, inception_5a_3_3, inception_5a_5_5, inception_5a_pool_1_1], axis=3,mode='concat')
inception_5b_1_1 = conv_2d(inception_5a_output, 384, filter_size=1,activation='relu', name='inception_5b_1_1')
inception_5b_3_3_reduce = conv_2d(inception_5a_output, 192, filter_size=1, activation='relu', name='inception_5b_3_3_reduce')
inception_5b_3_3 = conv_2d(inception_5b_3_3_reduce, 384, filter_size=3,activation='relu', name='inception_5b_3_3')
inception_5b_5_5_reduce = conv_2d(inception_5a_output, 48, filter_size=1, activation='relu', name='inception_5b_5_5_reduce')
inception_5b_5_5 = conv_2d(inception_5b_5_5_reduce,128, filter_size=5, activation='relu', name='inception_5b_5_5' )
inception_5b_pool = max_pool_2d(inception_5a_output, kernel_size=3, strides=1, name='inception_5b_pool')
inception_5b_pool_1_1 = conv_2d(inception_5b_pool, 128, filter_size=1, activation='relu', name='inception_5b_pool_1_1')
inception_5b_output = merge([inception_5b_1_1, inception_5b_3_3, inception_5b_5_5, inception_5b_pool_1_1], axis=3, mode='concat')
pool5_7_7 = avg_pool_2d(inception_5b_output, kernel_size=7, strides=1)
pool5_7_7 = dropout(pool5_7_7, 0.4)
loss = fully_connected(pool5_7_7, output,activation='softmax')
network = regression(loss, optimizer='momentum',
loss='categorical_crossentropy',
learning_rate=lr, name='targets')
model = tflearn.DNN(network,
max_checkpoints=0, tensorboard_verbose=0,tensorboard_dir='log')
return model
# IMDB Dataset loading
train, test, _ = imdb.load_data(path='imdb.pkl', n_words=10000,
valid_portion=0.1)
trainX, trainY = train
testX, testY = test
# Data preprocessing
# Sequence padding
trainX = pad_sequences(trainX, maxlen=100, value=0.)
testX = pad_sequences(testX, maxlen=100, value=0.)
# Converting labels to binary vectors
trainY = to_categorical(trainY)
testY = to_categorical(testY)
# Building convolutional network
network = input_data(shape=[None, 100], name='input')
network = tflearn.embedding(network, input_dim=10000, output_dim=128)
branch1 = conv_1d(network, 128, 3, padding='valid', activation='relu', regularizer="L2")
branch2 = conv_1d(network, 128, 4, padding='valid', activation='relu', regularizer="L2")
branch3 = conv_1d(network, 128, 5, padding='valid', activation='relu', regularizer="L2")
network = merge([branch1, branch2, branch3], mode='concat', axis=1)
network = tf.expand_dims(network, 2)
network = global_max_pool(network)
network = dropout(network, 0.5)
network = fully_connected(network, 2, activation='softmax')
network = regression(network, optimizer='adam', learning_rate=0.001,
loss='categorical_crossentropy', name='target')
# Training
model = tflearn.DNN(network, tensorboard_verbose=0)
model.fit(trainX, trainY, n_epoch = 5, shuffle=True, validation_set=(testX, testY), show_metric=True, batch_size=32)
pool1_3_3 = max_pool_2d(conv1_7_7, 3, strides=2) pool1_3_3 = local_response_normalization(pool1_3_3) conv2_3_3_reduce = conv_2d(pool1_3_3, 64, 1, activation='relu', name='conv2_3_3_reduce') conv2_3_3 = conv_2d(conv2_3_3_reduce, 192, 3, activation='relu', name='conv2_3_3') conv2_3_3 = local_response_normalization(conv2_3_3) pool2_3_3 = max_pool_2d(conv2_3_3, kernel_size=3, strides=2, name='pool2_3_3_s2') # 3a inception_3a_1_1 = conv_2d(pool2_3_3, 64, 1, activation='relu', name='inception_3a_1_1') inception_3a_3_3_reduce = conv_2d(pool2_3_3, 96, 1, activation='relu', name='inception_3a_3_3_reduce') inception_3a_3_3 = conv_2d(inception_3a_3_3_reduce, 128, filter_size=3, activation='relu', name='inception_3a_3_3') inception_3a_5_5_reduce = conv_2d(pool2_3_3, 16, filter_size=1, activation='relu', name='inception_3a_5_5_reduce') inception_3a_5_5 = conv_2d(inception_3a_5_5_reduce, 32, filter_size=5, activation='relu', name='inception_3a_5_5') inception_3a_pool = max_pool_2d(pool2_3_3, kernel_size=3, strides=1, name='inception_3a_pool') inception_3a_pool_1_1 = conv_2d(inception_3a_pool, 32, filter_size=1, activation='relu', name='inception_3a_pool_1_1') inception_3a_output = merge([inception_3a_1_1, inception_3a_3_3, inception_3a_5_5, inception_3a_pool_1_1], mode='concat', axis=3) # 3b inception_3b_1_1 = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_1_1') inception_3b_3_3_reduce = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_3_3_reduce') inception_3b_3_3 = conv_2d(inception_3b_3_3_reduce, 192, filter_size=3, activation='relu', name='inception_3b_3_3') inception_3b_5_5_reduce = conv_2d(inception_3a_output, 32, filter_size=1, activation='relu', name='inception_3b_5_5_reduce') inception_3b_5_5 = conv_2d(inception_3b_5_5_reduce, 96, filter_size=5, name='inception_3b_5_5') inception_3b_pool = max_pool_2d(inception_3a_output, kernel_size=3, strides=1, name='inception_3b_pool') inception_3b_pool_1_1 = conv_2d(inception_3b_pool, 64, filter_size=1, activation='relu', name='inception_3b_pool_1_1') inception_3b_output = merge([inception_3b_1_1, inception_3b_3_3, inception_3b_5_5, inception_3b_pool_1_1], mode='concat', axis=3, name='inception_3b_output') pool3_3_3 = max_pool_2d(inception_3b_output, kernel_size=3, strides=2, name='pool3_3_3') # 4a inception_4a_1_1 = conv_2d(pool3_3_3, 192, filter_size=1, activation='relu', name='inception_4a_1_1') inception_4a_3_3_reduce = conv_2d(pool3_3_3, 96, filter_size=1, activation='relu', name='inception_4a_3_3_reduce')
def construct_inceptionv1onfire (x,y):
# Build network as per architecture in [Dunnings/Breckon, 2018]
network = input_data(shape=[None, y, x, 3])
conv1_7_7 = conv_2d(network, 64, 5, strides=2, activation='relu', name = 'conv1_7_7_s2')
pool1_3_3 = max_pool_2d(conv1_7_7, 3,strides=2)
pool1_3_3 = local_response_normalization(pool1_3_3)
conv2_3_3_reduce = conv_2d(pool1_3_3, 64,1, activation='relu',name = 'conv2_3_3_reduce')
conv2_3_3 = conv_2d(conv2_3_3_reduce, 128,3, activation='relu', name='conv2_3_3')
conv2_3_3 = local_response_normalization(conv2_3_3)
pool2_3_3 = max_pool_2d(conv2_3_3, kernel_size=3, strides=2, name='pool2_3_3_s2')
inception_3a_1_1 = conv_2d(pool2_3_3, 64, 1, activation='relu', name='inception_3a_1_1')
inception_3a_3_3_reduce = conv_2d(pool2_3_3, 96,1, activation='relu', name='inception_3a_3_3_reduce')
inception_3a_3_3 = conv_2d(inception_3a_3_3_reduce, 128,filter_size=3, activation='relu', name = 'inception_3a_3_3')
inception_3a_5_5_reduce = conv_2d(pool2_3_3,16, filter_size=1,activation='relu', name ='inception_3a_5_5_reduce' )
inception_3a_5_5 = conv_2d(inception_3a_5_5_reduce, 32, filter_size=5, activation='relu', name= 'inception_3a_5_5')
inception_3a_pool = max_pool_2d(pool2_3_3, kernel_size=3, strides=1, )
inception_3a_pool_1_1 = conv_2d(inception_3a_pool, 32, filter_size=1, activation='relu', name='inception_3a_pool_1_1')
# merge the inception_3a__
inception_3a_output = merge([inception_3a_1_1, inception_3a_3_3, inception_3a_5_5, inception_3a_pool_1_1], mode='concat', axis=3)
inception_3b_1_1 = conv_2d(inception_3a_output, 128,filter_size=1,activation='relu', name= 'inception_3b_1_1' )
inception_3b_3_3_reduce = conv_2d(inception_3a_output, 128, filter_size=1, activation='relu', name='inception_3b_3_3_reduce')
inception_3b_3_3 = conv_2d(inception_3b_3_3_reduce, 192, filter_size=3, activation='relu',name='inception_3b_3_3')
inception_3b_5_5_reduce = conv_2d(inception_3a_output, 32, filter_size=1, activation='relu', name = 'inception_3b_5_5_reduce')
inception_3b_5_5 = conv_2d(inception_3b_5_5_reduce, 96, filter_size=5, name = 'inception_3b_5_5')
inception_3b_pool = max_pool_2d(inception_3a_output, kernel_size=3, strides=1, name='inception_3b_pool')
inception_3b_pool_1_1 = conv_2d(inception_3b_pool, 64, filter_size=1,activation='relu', name='inception_3b_pool_1_1')
#merge the inception_3b_*
inception_3b_output = merge([inception_3b_1_1, inception_3b_3_3, inception_3b_5_5, inception_3b_pool_1_1], mode='concat',axis=3,name='inception_3b_output')
pool3_3_3 = max_pool_2d(inception_3b_output, kernel_size=3, strides=2, name='pool3_3_3')
inception_4a_1_1 = conv_2d(pool3_3_3, 192, filter_size=1, activation='relu', name='inception_4a_1_1')
inception_4a_3_3_reduce = conv_2d(pool3_3_3, 96, filter_size=1, activation='relu', name='inception_4a_3_3_reduce')
inception_4a_3_3 = conv_2d(inception_4a_3_3_reduce, 208, filter_size=3, activation='relu', name='inception_4a_3_3')
inception_4a_5_5_reduce = conv_2d(pool3_3_3, 16, filter_size=1, activation='relu', name='inception_4a_5_5_reduce')
inception_4a_5_5 = conv_2d(inception_4a_5_5_reduce, 48, filter_size=5, activation='relu', name='inception_4a_5_5')
inception_4a_pool = max_pool_2d(pool3_3_3, kernel_size=3, strides=1, name='inception_4a_pool')
inception_4a_pool_1_1 = conv_2d(inception_4a_pool, 64, filter_size=1, activation='relu', name='inception_4a_pool_1_1')
inception_4a_output = merge([inception_4a_1_1, inception_4a_3_3, inception_4a_5_5, inception_4a_pool_1_1], mode='concat', axis=3, name='inception_4a_output')
pool5_7_7 = avg_pool_2d(inception_4a_output, kernel_size=5, strides=1)
pool5_7_7 = dropout(pool5_7_7, 0.4)
loss = fully_connected(pool5_7_7, 2,activation='softmax')
network = regression(loss, optimizer='momentum',
loss='categorical_crossentropy',
learning_rate=0.001)
model = tflearn.DNN(network, checkpoint_path='inceptiononv1onfire',
max_checkpoints=1, tensorboard_verbose=2)
return model
