Functional model

One 、 Build a single channel model

Build a single channel model , Use sequential （Sequential） Models and functions （Functional） The model can be built .

1. sequential （Sequential） Model structures,

Build code ：

#  Define sequential models 
model = Sequential()

#  The first convolution layer 
# input_shape  Input plane 
# filters  Convolution kernel / Number of filters 
# kernel_size  Convolution window size 
# strides  step 
# padding padding The way  same/valid
# activation  Activation function 
model.add(Conv2D(
    input_shape = (28,28,1),
    filters = 32,
    kernel_size = 5,
    strides = 1,
    padding = 'same',
    activation = 'relu'
))
#  The first pool 
model.add(MaxPooling2D(
    pool_size = 2,
    strides = 2,
    padding = 'same',
))
#  Second convolution layer 
model.add(Conv2D(64,5,strides=1,padding='same',activation = 'relu'))
#  The second pooling layer 
model.add(MaxPooling2D(2,2,'same'))
#  Flatten the output of the second pool layer into 1 dimension 
model.add(Flatten())
#  The first full connection layer 
model.add(Dense(1024,activation = 'relu'))
# Dropout
model.add(Dropout(0.5))
#  The second full connection layer 
model.add(Dense(10,activation='softmax'))

2. Functional expression （Functional） Model structures,

Build code ：

inputs = Input(shape=(28,28,1))
x = Conv2D(filters=32, kernel_size=5, padding='same', activation='relu')(inputs)
x = MaxPooling2D(pool_size = 2)(x)
x = Conv2D(filters=64, kernel_size=5, padding='same', activation='relu')(x)
x = MaxPooling2D(pool_size = 2)(x)
x = Flatten()(x)
x = Dense(1024,activation = 'relu')(x)
x = Dropout(0.5)(x)
predictions = Dense(10,activation='softmax')(x)
model = Model(inputs=inputs, outputs=predictions)

You need to import from keras.models import Model.

Two 、 Build a multi-channel model

In addition to the common single channel model , There are also many well-known multi-channel models . for example google Of Inception:

It is a typical multi-channel model . sequential （Sequential） A model can only define a model with one channel , If a model has many channels , that Sequential() Sequential models cannot be defined , At this time, only functional models can be used to build network models .
Build some multi-channel Inception Network model code ：

from keras.layers import Conv2D, MaxPooling2D, Input

input_img = Input(shape=(256, 256, 3))

tower_1 = Conv2D(64, (1, 1), padding='same', activation='relu')(input_img)
tower_1 = Conv2D(64, (3, 3), padding='same', activation='relu')(tower_1)

tower_2 = Conv2D(64, (1, 1), padding='same', activation='relu')(input_img)
tower_2 = Conv2D(64, (5, 5), padding='same', activation='relu')(tower_2)

tower_3 = MaxPooling2D((3, 3), strides=(1, 1), padding='same')(input_img)
tower_3 = Conv2D(64, (1, 1), padding='same', activation='relu')(tower_3)

output = keras.layers.concatenate([tower_1, tower_2, tower_3], axis=1)

3、 ... and 、 Complete code

The code running platform is jupyter-notebook, Code blocks in the article , According to jupyter-notebook Written in the order of division in , Run article code , Glue directly into jupyter-notebook that will do .

1. Import third-party library

import numpy as np
from keras.datasets import mnist
from keras.utils import np_utils
from keras.models import Sequential
from keras.layers import Dense,Dropout,Conv2D,MaxPooling2D,Flatten,Input
from keras.optimizers import Adam
from keras.models import Model

2. Data processing

#  Load data 
(x_train,y_train),(x_test,y_test) = mnist.load_data()
# (60000,28,28)->(60000,28,28,1)
x_train = x_train.reshape(-1,28,28,1)/255.0
x_test = x_test.reshape(-1,28,28,1)/255.0
#  in one hot Format 
y_train = np_utils.to_categorical(y_train,num_classes=10)
y_test = np_utils.to_categorical(y_test,num_classes=10)

3. sequential （Sequential） Model building a single channel network

#  Define sequential models 
model = Sequential()

#  The first convolution layer 
# input_shape  Input plane 
# filters  Convolution kernel / Number of filters 
# kernel_size  Convolution window size 
# strides  step 
# padding padding The way  same/valid
# activation  Activation function 
model.add(Conv2D(
    input_shape = (28,28,1),
    filters = 32,
    kernel_size = 5,
    strides = 1,
    padding = 'same',
    activation = 'relu'
))
#  The first pool 
model.add(MaxPooling2D(
    pool_size = 2,
    strides = 2,
    padding = 'same',
))
#  Second convolution layer 
model.add(Conv2D(64,5,strides=1,padding='same',activation = 'relu'))
#  The second pooling layer 
model.add(MaxPooling2D(2,2,'same'))
#  Flatten the output of the second pool layer into 1 dimension 
model.add(Flatten())
#  The first full connection layer 
model.add(Dense(1024,activation = 'relu'))
# Dropout
model.add(Dropout(0.5))
#  The second full connection layer 
model.add(Dense(10,activation='softmax'))

4. Functional expression （Functional） Model building a single channel network

inputs = Input(shape=(28,28,1))
x = Conv2D(filters=32, kernel_size=5, padding='same', activation='relu')(inputs)
x = MaxPooling2D(pool_size = 2)(x)
x = Conv2D(filters=64, kernel_size=5, padding='same', activation='relu')(x)
x = MaxPooling2D(pool_size = 2)(x)
x = Flatten()(x)
x = Dense(1024,activation = 'relu')(x)
x = Dropout(0.5)(x)
predictions = Dense(10,activation='softmax')(x)
model = Model(inputs=inputs, outputs=predictions)

5. Training models

#  Define optimizer 
adam = Adam(lr=1e-4)

#  Define optimizer ,loss function, The accuracy of calculation during training 
model.compile(optimizer=adam,loss='categorical_crossentropy',metrics=['accuracy'])

#  Training models 
model.fit(x_train,y_train,batch_size=64,epochs=10)

#  Evaluation model 
loss,accuracy = model.evaluate(x_test,y_test)

print('test loss',loss)
print('test accuracy',accuracy)

6. Functional model to build a multi-channel network

from keras.layers import Conv2D, MaxPooling2D, Input

input_img = Input(shape=(256, 256, 3))

tower_1 = Conv2D(64, (1, 1), padding='same', activation='relu')(input_img)
tower_1 = Conv2D(64, (3, 3), padding='same', activation='relu')(tower_1)

tower_2 = Conv2D(64, (1, 1), padding='same', activation='relu')(input_img)
tower_2 = Conv2D(64, (5, 5), padding='same', activation='relu')(tower_2)

tower_3 = MaxPooling2D((3, 3), strides=(1, 1), padding='same')(input_img)
tower_3 = Conv2D(64, (1, 1), padding='same', activation='relu')(tower_3)

output = keras.layers.concatenate([tower_1, tower_2, tower_3], axis=1)