tensorflow 2.0 keras 高层接口 之 自定义层网络

Outline

• keras.Sequential
• Keras.layers.Layer
• Keras.Model

注:

1. 要想使用 Sequential 必须遵循一些协议。
   • 自定的层继承至 Keras.layers.Layer 类。
   • 自己的模型也必须要继承至 Keras.Model 类 。

keras.Sequential

network = Sequential([layers.Dense(256, activation='relu'),
                      layers.Dense(128, activation='relu'),
                      layers.Dense(64, activation='relu'),
                      layers.Dense(32, activation='relu'),
                      layers.Dense(10)])
network.build(input_shape=(None, 28 * 28))   # 建立w和b参数
network.summary()   # 打印网络参数

解释：

1. Sequential 容器内的类全部继承至 Keras.layers.Layer 类 。
2. Sequential 还可以方便管理参数。
   • model.trainable_variables # 打印参数信息
   • model.call() # 向前传递
     • model(x) 是调用 model.__call__(x)方法，再调用 call(x) 方法。

Layer/Model

1. 继承 Keras.layers.Layer 和 keras.Model (两个自定义层模板母类)
2. 实现 __init__ (super 继承 母类 的初始化方法， 额外通用属性）
3. 实现 call （实现自己的逻辑）
4. Model：compile/fit/evaluate

自定义层

class MyDense(layers.Layer):

    def __init__(self, inp_dim, outp_dim):
        super(MyDense, self).__init__()

        self.kernel = self.add_variable('w', [inp_dim, outp_dim])   # 使用 add_variable(母类方法) 创建可训练变量，交给模型管理。
        self.bias = self.add_variable('b', [outp_dim])

    def call(self, inputs, training=None):   # 训练测试关键字 train 
        out = inputs @ self.kernel + self.bias

        return out

自定义网络结构

class MyModel(keras.Model):

    def __init__(self):
        super(MyModel, self).__init__()

        self.fc1 = MyDense(28 * 28, 256)
        self.fc2 = MyDense(256, 128)
        self.fc3 = MyDense(128, 64)
        self.fc4 = MyDense(64, 32)
        self.fc5 = MyDense(32, 10)

    def call(self, inputs, training=None):
        x = self.fc1(inputs)
        x = tf.nn.relu(x)
        x = self.fc2(x)
        x = tf.nn.relu(x)
        x = self.fc3(x)
        x = tf.nn.relu(x)
        x = self.fc4(x)
        x = tf.nn.relu(x)
        x = self.fc5(x)

        return x

完整代码

import os
import tensorflow as tf
from tensorflow.keras import datasets, layers, optimizers, Sequential, metrics
from tensorflow import keras

os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'


def preprocess(x, y):
    """
    x is a simple image, not a batch
    """
    x = tf.cast(x, dtype=tf.float32) / 255.
    x = tf.reshape(x, [28 * 28])
    y = tf.cast(y, dtype=tf.int32)
    y = tf.one_hot(y, depth=10)
    return x, y


batchsz = 128
(x, y), (x_val, y_val) = datasets.mnist.load_data()
print('datasets:', x.shape, y.shape, x.min(), x.max())

db = tf.data.Dataset.from_tensor_slices((x, y))
db = db.map(preprocess).shuffle(60000).batch(batchsz)
ds_val = tf.data.Dataset.from_tensor_slices((x_val, y_val))
ds_val = ds_val.map(preprocess).batch(batchsz)

sample = next(iter(db))
print(sample[0].shape, sample[1].shape)

network = Sequential([layers.Dense(256, activation='relu'),
                      layers.Dense(128, activation='relu'),
                      layers.Dense(64, activation='relu'),
                      layers.Dense(32, activation='relu'),
                      layers.Dense(10)])
network.build(input_shape=(None, 28 * 28))
network.summary()




class MyDense(layers.Layer):

    def __init__(self, inp_dim, outp_dim):
        super(MyDense, self).__init__()

        self.kernel = self.add_variable('w', [inp_dim, outp_dim])
        self.bias = self.add_variable('b', [outp_dim])

    def call(self, inputs, training=None):
        out = inputs @ self.kernel + self.bias

        return out


class MyModel(keras.Model):

    def __init__(self):
        super(MyModel, self).__init__()

        self.fc1 = MyDense(28 * 28, 256)
        self.fc2 = MyDense(256, 128)
        self.fc3 = MyDense(128, 64)
        self.fc4 = MyDense(64, 32)
        self.fc5 = MyDense(32, 10)

    def call(self, inputs, training=None):
        x = self.fc1(inputs)
        x = tf.nn.relu(x)
        x = self.fc2(x)
        x = tf.nn.relu(x)
        x = self.fc3(x)
        x = tf.nn.relu(x)
        x = self.fc4(x)
        x = tf.nn.relu(x)
        x = self.fc5(x)

        return x


network = MyModel()

network.compile(optimizer=optimizers.Adam(lr=0.01),
                loss=tf.losses.CategoricalCrossentropy(from_logits=True),
                metrics=['accuracy']
                )

network.fit(db, epochs=5, validation_data=ds_val,
            validation_freq=2)

network.evaluate(ds_val)

sample = next(iter(ds_val))
x = sample[0]
y = sample[1]  # one-hot
pred = network.predict(x)  # [b, 10]
# convert back to number
y = tf.argmax(y, axis=1)
pred = tf.argmax(pred, axis=1)

print(pred)
print(y)