Tensorflow Keras

翻译自：https://www.tensorflow.org/guide/keras

导入 tf.keras

import tensorflow as tf
from tensorflow import keras

通过 Sequential model 构建神经网络

model = keras.Sequential()
# Adds a densely-connected layer with 64 units to the model:
model.add(keras.layers.Dense(64, activation='relu'))
# Add another:
model.add(keras.layers.Dense(64, activation='relu'))
# Add a softmax layer with 10 output units:
model.add(keras.layers.Dense(10, activation='softmax'))

针对每一层可以给出激活函数，正则化方法和初始化方法

# Create a sigmoid layer:
layers.Dense(64, activation='sigmoid')
# Or:
layers.Dense(64, activation=tf.sigmoid)

# A linear layer with L1 regularization of factor 0.01 applied to the kernel matrix:
layers.Dense(64, kernel_regularizer=keras.regularizers.l1(0.01))
# A linear layer with L2 regularization of factor 0.01 applied to the bias vector:
layers.Dense(64, bias_regularizer=keras.regularizers.l2(0.01))

# A linear layer with a kernel initialized to a random orthogonal matrix:
layers.Dense(64, kernel_initializer='orthogonal')
# A linear layer with a bias vector initialized to 2.0s:
layers.Dense(64, bias_initializer=keras.initializers.constant(2.0))

训练

# Configure a model for mean-squared error regression.
model.compile(optimizer=tf.train.AdamOptimizer(0.01),
              loss='mse',       # mean squared error
              metrics=['mae'])  # mean absolute error

# Configure a model for categorical classification.
model.compile(optimizer=tf.train.RMSPropOptimizer(0.01),
              loss=keras.losses.categorical_crossentropy,
              metrics=[keras.metrics.categorical_accuracy])

• optimizer: 给出参数的优化方法，主要包括 AdamOptimizer, RMSPropOptimizer 和 GradientDescentOptimizer
• loss：给出代价函数或者称为目标函数，主要包括 mse (均方误差，用于回归)，categorical_crossentropy 和 binary_crossentropy（用于概率分类）
• metrics：用于监控训练过程，主要包括 mae (mean absolute error，用于回归)，accuracy, binary_crossentropy

model.fit(data, labels, epochs=10, batch_size=32)

model.fit(data, labels, epochs=10, batch_size=32,
          validation_data=(val_data, val_labels))

# Instantiates a toy dataset instance:
dataset = tf.data.Dataset.from_tensor_slices((data, labels))
dataset = dataset.batch(32)
dataset = dataset.repeat()

# Don't forget to specify `steps_per_epoch` when calling `fit` on a dataset.
model.fit(dataset, epochs=10, steps_per_epoch=30)

dataset = tf.data.Dataset.from_tensor_slices((data, labels))
dataset = dataset.batch(32).repeat()

val_dataset = tf.data.Dataset.from_tensor_slices((val_data, val_labels))
val_dataset = val_dataset.batch(32).repeat()

model.fit(dataset, epochs=10, steps_per_epoch=30,
          validation_data=val_dataset,
          validation_steps=3)

评估和预测

model.evaluate(x, y, batch_size=32)

model.evaluate(dataset, steps=30)

model.predict(x, batch_size=32)

model.predict(dataset, steps=30)

创建复杂网络

inputs = keras.Input(shape=(32,))  # Returns a placeholder tensor

# A layer instance is callable on a tensor, and returns a tensor.
x = keras.layers.Dense(64, activation='relu')(inputs)
x = keras.layers.Dense(64, activation='relu')(x)
predictions = keras.layers.Dense(10, activation='softmax')(x)

# Instantiate the model given inputs and outputs.
model = keras.Model(inputs=inputs, outputs=predictions)

# The compile step specifies the training configuration.
model.compile(optimizer=tf.train.RMSPropOptimizer(0.001),
              loss='categorical_crossentropy',
              metrics=['accuracy'])

# Trains for 5 epochs
model.fit(data, labels, batch_size=32, epochs=5)

子类模型

class MyModel(keras.Model):

  def __init__(self, num_classes=10):
    super(MyModel, self).__init__(name='my_model')
    self.num_classes = num_classes
    # Define your layers here.
    self.dense_1 = keras.layers.Dense(32, activation='relu')
    self.dense_2 = keras.layers.Dense(num_classes, activation='sigmoid')

  def call(self, inputs):
    # Define your forward pass here,
    # using layers you previously defined (in `__init__`).
    x = self.dense_1(inputs)
    return self.dense_2(x)

  def compute_output_shape(self, input_shape):
    # You need to override this function if you want to use the subclassed model
    # as part of a functional-style model.
    # Otherwise, this method is optional.
    shape = tf.TensorShape(input_shape).as_list()
    shape[-1] = self.num_classes
    return tf.TensorShape(shape)


# Instantiates the subclassed model.
model = MyModel(num_classes=10)

# The compile step specifies the training configuration.
model.compile(optimizer=tf.train.RMSPropOptimizer(0.001),
              loss='categorical_crossentropy',
              metrics=['accuracy'])

# Trains for 5 epochs.
model.fit(data, labels, batch_size=32, epochs=5)

自定义层

class MyLayer(keras.layers.Layer):

  def __init__(self, output_dim, **kwargs):
    self.output_dim = output_dim
    super(MyLayer, self).__init__(**kwargs)

  def build(self, input_shape):
    shape = tf.TensorShape((input_shape[1], self.output_dim))
    # Create a trainable weight variable for this layer.
    self.kernel = self.add_weight(name='kernel',
                                  shape=shape,
                                  initializer='uniform',
                                  trainable=True)
    # Be sure to call this at the end
    super(MyLayer, self).build(input_shape)

  def call(self, inputs):
    return tf.matmul(inputs, self.kernel)

  def compute_output_shape(self, input_shape):
    shape = tf.TensorShape(input_shape).as_list()
    shape[-1] = self.output_dim
    return tf.TensorShape(shape)

  def get_config(self):
    base_config = super(MyLayer, self).get_config()
    base_config['output_dim'] = self.output_dim

  @classmethod
  def from_config(cls, config):
    return cls(**config)


# Create a model using the custom layer
model = keras.Sequential([MyLayer(10),
                          keras.layers.Activation('softmax')])

# The compile step specifies the training configuration
model.compile(optimizer=tf.train.RMSPropOptimizer(0.001),
              loss='categorical_crossentropy',
              metrics=['accuracy'])

# Trains for 5 epochs.
model.fit(data, targets, batch_size=32, epochs=5)

回调函数

callbacks = [
  # Interrupt training if `val_loss` stops improving for over 2 epochs
  keras.callbacks.EarlyStopping(patience=2, monitor='val_loss'),
  # Write TensorBoard logs to `./logs` directory
  keras.callbacks.TensorBoard(log_dir='./logs')
]
model.fit(data, labels, batch_size=32, epochs=5, callbacks=callbacks,
          validation_data=(val_data, val_targets))

主要的回调函数包括

• ModelCheckpoint：固定时间保存 checkpoint
• LearningRateScheduler：动态改变学习率
• EarlyStopping：当交叉验证集性能下降时，中断训练
• TensorBoard：使用 TensorBoard 监控模型 （tensorboard --logdir /tmp/retrain_logs）

模型的保存与恢复

只保存权重

# Save weights to a TensorFlow Checkpoint file
model.save_weights('./my_model')

# Restore the model's state,
# this requires a model with the same architecture.
model.load_weights('my_model')

保存权重为 Keras HDF5 格式 

# Save weights to a HDF5 file
model.save_weights('my_model.h5', save_format='h5')

# Restore the model's state
model.load_weights('my_model.h5')

只保存配置

# Serialize a model to JSON format
json_string = model.to_json()

# Recreate the model (freshly initialized)
fresh_model = keras.models.from_json(json_string)

# Serializes a model to YAML format
yaml_string = model.to_yaml()

# Recreate the model
fresh_model = keras.models.from_yaml(yaml_string)

保存整个模型

# Create a trivial model
model = keras.Sequential([
  keras.layers.Dense(10, activation='softmax', input_shape=(32,)),
  keras.layers.Dense(10, activation='softmax')
])
model.compile(optimizer='rmsprop',
              loss='categorical_crossentropy',
              metrics=['accuracy'])
model.fit(data, targets, batch_size=32, epochs=5)


# Save entire model to a HDF5 file
model.save('my_model.h5')

# Recreate the exact same model, including weights and optimizer.
model = keras.models.load_model('my_model.h5')

多 GPU

model = keras.Sequential()
model.add(keras.layers.Dense(16, activation='relu', input_shape=(10,)))
model.add(keras.layers.Dense(1, activation='sigmoid'))

optimizer = tf.train.GradientDescentOptimizer(0.2)

model.compile(loss='binary_crossentropy', optimizer=optimizer)
model.summary()

def input_fn():
  x = np.random.random((1024, 10))
  y = np.random.randint(2, size=(1024, 1))
  x = tf.cast(x, tf.float32)
  dataset = tf.data.Dataset.from_tensor_slices((x, y))
  dataset = dataset.repeat(10)
  dataset = dataset.batch(32)
  return dataset

strategy = tf.contrib.distribute.MirroredStrategy()
config = tf.estimator.RunConfig(train_distribute=strategy)

keras_estimator = keras.estimator.model_to_estimator(
  keras_model=model,
  config=config,
  model_dir='/tmp/model_dir')

keras_estimator.train(input_fn=input_fn, steps=10)