SciTech-BigDataAIML-Tensorflow-Keras的API列表 + TensorFlow 模型建立与训练

Tensorflow链接:

https://www.tensorflow.org/install

https://www.tensorflow.org/guide

https://www.tensorflow.org/tutorials

https://www.tensorflow.org/learn

Tensorflow模型介绍：

https://tf.wiki/zh_hans/basic/models.html

Keras API查询列表：

https://keras.io/api/layers/core_layers/embedding/

https://keras.io/api/layers/reshaping_layers/flatten/

模型（Model）与层（Layer）

TensorFlow 推荐使用 Keras(tf.keras) 构建模型。

Keras 是一个广为流行的高级神经网络 API，简单、快速而不失灵活性，现已得到 TensorFlow 的官方内置和全面支持。

Keras 有两个重要的概念:Model(模型) 和 Layer(层) 。

• Layer层将各种计算流程和变量进行封装(例如全连接层，CNN 卷积层和池化层等);
• Model模型则将各种Layer层进行组织和连接，并封装成一个整体，描述了如何将输入数据通过各种层以及运算而得到输出。

调用模型，使用 y_pred = model(X) 的形式即可。

Keras 在 tf.keras.layers 处内置深度学习会大量常用的的预定义层，同时也允许我们自定义层。

Keras 的Model模型以class类的形式呈现，我们可以通过继承 tf.keras.Model 这个 Python 类来定义自己的模型。要继承类，需要重写 init() （构造函数，初始化）和 call(input) （模型调用）两个方法，同时也可以根据需要增加自定义的方法。

Tensorflow Models API

There are three ways to create Keras models:

• The Sequential model, which is very straightforward (a simple list of layers), but is limited to single-input, single-output stacks of layers (as the name gives away).
• The Functional API, which is an easy-to-use, fully-featured API that supports arbitrary model architectures. For most people and most use cases, this is what you should be using. This is the Keras "industry strength" model.
• Model subclassing, where you implement everything from scratch on your own. Use this if you have complex, out-of-the-box research use cases.

Models API overview

The Model class

Model class

summary method

get_layer method

The Sequential class

Sequential class

add method

pop method

Model training APIs

compile method

fit method

evaluate method

predict method

train_on_batch method

test_on_batch method

predict_on_batch method

Saving & serialization

Whole model saving & loading

Weights-only saving & loading

Model config serialization

Model export for inference

Serialization utilities

Source Code Example

import tensorflow as tf

mnist = tf.keras.datasets.mnist

(x_train, y_train),(x_test, y_test) = mnist.load_data()

x_train, x_test = x_train / 255.0, x_test / 255.0

顺序模型：

model = tf.keras.models.Sequential([

tf.keras.layers.Flatten(input_shape=(28, 28)),

tf.keras.layers.Dense(128, activation='relu'),

tf.keras.layers.Dropout(0.2),

tf.keras.layers.Dense(10, activation='softmax')

])

model.compile(optimizer='adam',

loss='sparse_categorical_crossentropy',

metrics=['accuracy'])

model.fit(x_train, y_train, epochs=5)

model.evaluate(x_test, y_test)