Keras深度学习模型Sequential和Model详解

# 导入类
from keras.models import Sequential
from keras.layers import Dense, Activation

# 构建Sequential模型
# Model是keras最核心的数据结构
model = Sequential([
    Dense(32, input_shape=(784,)),
    Activation('relu'),
    Dense(10),
    Activation('softmax'),
])

model = Sequential()
model.add(Dense(32, input_dim=784))
model.add(Activation('relu'))
model.add(Dense(10))
model.add(Activation('softmax'))

model.add(Dense(32, input_shape=(784,))

model.add(Dense(32, input_dim=784))

# 分类问题
model.compile(
    optimizer='rmsprop',
    loss='categorical_crossentropy',
    metrics=['accuracy']
)
# 二分类问题
model.compile(
    optimizer='rmsprop',
    loss='binary_crossentropy',
    metrics=['accuracy']
)
# 均方误差回归问题
model.compile(
    optimizer='rmsprop',
    loss='mse'
)
# 自定义评估标准函数
import keras.backend as K
def mean_pred(y_true, y_pred):
    return K.mean(y_pred)
model.compile(
    optimizer='rmsprop',
    loss='binary_crossentropy',
    metrics=['accuracy', mean_pred]
)

# 对于具有2个类的单输入模型（二进制分类）
model = Sequential()
model.add(Dense(32, activation='relu', input_dim=100))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='rmsprop',
             loss='binary_crossentropy',
             metrics=['accuracy'])
# 生成虚拟数据
import numpy as np
data = np.random.random((1000, 100))
labels = np.random.randint(2, size=(1000, 1))
# 训练模型，以32个样本为一个batch进行迭代
model.fit(data, labels, epochs=10, batch_size=32)

Epoch 1/10
1000/1000 [==============================] - 0s 105us/step - loss: 0.7028 - accuracy: 0.4980
Epoch 2/10
1000/1000 [==============================] - 0s 32us/step - loss: 0.6932 - accuracy: 0.5380
Epoch 3/10
1000/1000 [==============================] - 0s 34us/step - loss: 0.6862 - accuracy: 0.5510
Epoch 4/10
1000/1000 [==============================] - 0s 34us/step - loss: 0.6842 - accuracy: 0.5580
Epoch 5/10
1000/1000 [==============================] - 0s 31us/step - loss: 0.6834 - accuracy: 0.5570
Epoch 6/10
1000/1000 [==============================] - 0s 34us/step - loss: 0.6799 - accuracy: 0.5720
Epoch 7/10
1000/1000 [==============================] - 0s 34us/step - loss: 0.6760 - accuracy: 0.5860
Epoch 8/10
1000/1000 [==============================] - 0s 37us/step - loss: 0.6742 - accuracy: 0.5920
Epoch 9/10
1000/1000 [==============================] - 0s 35us/step - loss: 0.6702 - accuracy: 0.5810
Epoch 10/10
1000/1000 [==============================] - 0s 36us/step - loss: 0.6686 - accuracy: 0.6050
<keras.callbacks.callbacks.History at 0x1e1c574b888>

from keras.layers import Input, Dense
from keras.models import Model
# 定义输入层，确定输入维度
input = input(shape = (784, ))
# 2个隐含层，每个都有64个神经元，使用relu激活函数，且由上一层作为参数
x = Dense(64, activation='relu')(input)
x = Dense(64, activation='relu')(x)
# 输出层
y = Dense(10, activation='softmax')(x)
# 定义模型，指定输入输出
model = Model(input=input, output=y)
# 编译模型，指定优化器，损失函数，度量
model.compile(optimizer='rmsprop', loss='categorical_crossentropy', metrics=['accuracy'])
# 模型拟合，即训练
model.fit(data, labels)