3 keras版本模型训练

顺序模型

from tensorflow.keras import layers
import tensorflow as tf
import numpy as np

data = np.random.random((1000, 32))
labels = np.random.random((1000, 10))

# 搭建模型
model = tf.keras.Sequential()
model.add(layers.Dense(64, activation='relu'))#第一层
model.add(layers.Dense(64, activation='relu'))#第二层
model.add(layers.Dense(10))#第三层

# 指定损失函数优化器那些
model.compile(optimizer=tf.keras.optimizers.Adam(0.01),
              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

# 回调函数
callbacks = [
    # 早停
    tf.keras.callbacks.EarlyStopping(
        # 当‘val_loss’不再下降时候停止训练 
        monitor='val_loss',
        # “不再下降”被定义为“减少不超过1e-2”
        min_delta=1e-2,
        # “不再改善”进一步定义为“至少2个epoch”
        patience=2,
        verbose=1),
    # 保存权重
    tf.keras.callbacks.ModelCheckpoint(
        filepath='mymodel_{epoch}',
        # 模型保存路径
        #
        # 下面的两个参数意味着当且仅当`val_loss`分数提高时，我们才会覆盖当前检查点。
        save_best_only=True,
        monitor='val_loss',
        #加入这个仅仅保存模型权重
        save_weights_only=True,
        verbose=1),
    # 动态调整学习率
    tf.keras.callbacks.ReduceLROnPlateau(monitor="val_loss", 
                                         verbose=1, 
                                         mode='max', 
                                         factor=0.5, 
                                         patience=3)
]

# 训练
model.fit(data, labels,
          epochs=30,
          batch_size=64,
          callbacks=callbacks,
          validation_split=0.2
         )

顺序模型 2

from tensorflow.keras import layers
import tensorflow as tf
import numpy as np

data = np.random.random((1000, 32))
labels = np.random.random((1000, 10))

# 搭建模型
model = tf.keras.Sequential([
    layers.Dense(64, activation='relu', input_shape=(32,)),#第一层
    layers.Dense(64, activation='relu'),#第二层
    layers.Dense(10)#第三层
])


# 指定损失函数优化器那些
model.compile(optimizer=tf.keras.optimizers.Adam(0.01),
              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

# 回调函数
callbacks = [
    # 早停
    tf.keras.callbacks.EarlyStopping(
        # 当‘val_loss’不再下降时候停止训练 
        monitor='val_loss',
        # “不再下降”被定义为“减少不超过1e-2”
        min_delta=1e-2,
        # “不再改善”进一步定义为“至少2个epoch”
        patience=2,
        verbose=1),
    # 保存权重
    tf.keras.callbacks.ModelCheckpoint(
        filepath='mymodel_{epoch}',
        # 模型保存路径
        #
        # 下面的两个参数意味着当且仅当`val_loss`分数提高时，我们才会覆盖当前检查点。
        save_best_only=True,
        monitor='val_loss',
        #加入这个仅仅保存模型权重
        save_weights_only=True,
        verbose=1),
    # 动态调整学习率
    tf.keras.callbacks.ReduceLROnPlateau(monitor="val_loss", 
                                         verbose=1, 
                                         mode='max', 
                                         factor=0.5, 
                                         patience=3)
]

# 训练
model.fit(data, labels,
          epochs=30,
          batch_size=64,
          callbacks=callbacks,
          validation_split=0.2
         )

函数式

from tensorflow.keras import layers
import tensorflow as tf
import numpy as np

data = np.random.random((1000, 32))
labels = np.random.random((1000, 10))

inputs = tf.keras.Input(shape=(32,))  
# inputs = tf.keras.Input(shape=(32,))  
x = layers.Dense(64, activation='relu')(inputs) #第一层
x = layers.Dense(64, activation='relu')(x) #第二层
predictions = layers.Dense(10)(x) #第三层
model = tf.keras.Model(inputs=inputs, outputs=predictions)

# 指定损失函数优化器那些
model.compile(optimizer=tf.keras.optimizers.Adam(0.01),
              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

# 回调函数
callbacks = [
    # 早停
    tf.keras.callbacks.EarlyStopping(
        # 当‘val_loss’不再下降时候停止训练 
        monitor='val_loss',
        # “不再下降”被定义为“减少不超过1e-2”
        min_delta=1e-2,
        # “不再改善”进一步定义为“至少2个epoch”
        patience=2,
        verbose=1),
    # 保存权重
    tf.keras.callbacks.ModelCheckpoint(
        filepath='mymodel_{epoch}',
        # 模型保存路径
        #
        # 下面的两个参数意味着当且仅当`val_loss`分数提高时，我们才会覆盖当前检查点。
        save_best_only=True,
        monitor='val_loss',
        #加入这个仅仅保存模型权重
        save_weights_only=True,
        verbose=1),
    # 动态调整学习率
    tf.keras.callbacks.ReduceLROnPlateau(monitor="val_loss", 
                                         verbose=1, 
                                         mode='max', 
                                         factor=0.5, 
                                         patience=3)
]

# 训练
model.fit(data, labels,
          epochs=30,
          batch_size=64,
          callbacks=callbacks,
          validation_split=0.2
         )

子类化模型

from tensorflow.keras import layers
import tensorflow as tf
import numpy as np

data = np.random.random((1000, 32))
labels = np.random.random((1000, 10))

class MyModel(tf.keras.Model):

    def __init__(self, num_classes=10):
        super(MyModel, self).__init__(name='my_model')
        self.num_classes = num_classes
        # 定义自己需要的层
        self.dense_1 = layers.Dense(32, activation='relu') #
        self.dense_2 = layers.Dense(num_classes)

    def call(self, inputs):
        #定义前向传播
        # 使用在 (in `__init__`)定义的层
        x = self.dense_1(inputs)
        x = self.dense_2(x)
        return x
    
model = MyModel(num_classes=10)

# 指定损失函数优化器那些
model.compile(optimizer=tf.keras.optimizers.Adam(0.01),
              loss=tf.keras.losses.CategoricalCrossentropy(from_logits=True),
              metrics=['accuracy'])

# 回调函数
callbacks = [
    # 早停
    tf.keras.callbacks.EarlyStopping(
        # 当‘val_loss’不再下降时候停止训练 
        monitor='val_loss',
        # “不再下降”被定义为“减少不超过1e-2”
        min_delta=1e-2,
        # “不再改善”进一步定义为“至少2个epoch”
        patience=2,
        verbose=1),
    # 保存权重
    tf.keras.callbacks.ModelCheckpoint(
        filepath='mymodel_{epoch}',
        # 模型保存路径
        #
        # 下面的两个参数意味着当且仅当`val_loss`分数提高时，我们才会覆盖当前检查点。
        save_best_only=True,
        monitor='val_loss',
        #加入这个仅仅保存模型权重
        save_weights_only=True,
        verbose=1),
    # 动态调整学习率
    tf.keras.callbacks.ReduceLROnPlateau(monitor="val_loss", 
                                         verbose=1, 
                                         mode='max', 
                                         factor=0.5, 
                                         patience=3)
]

# 训练
model.fit(data, labels,
          epochs=30,
          batch_size=64,
          callbacks=callbacks,
          validation_split=0.2
         )

画图

tf.keras.utils.plot_model(model, 'multi_input_and_output_model.png', show_shapes=True,dpi=500)

模型训练：model.fit()
模型验证：model.evaluate()
模型预测： model.predict()

# Evaluate the model on the test data using `evaluate`
print('\n# Evaluate on test data')
results = model.evaluate(x_test, y_test, batch_size=128)
print('test loss, test acc:', results)

# Generate predictions (probabilities -- the output of the last layer)
# on new data using `predict`
print('\n# Generate predictions for 3 samples')
predictions = model.predict(x_test[:3])
print('predictions shape:', predictions.shape)