TensorFlow 2 Pocket Reference

Index

A

• AI Fairness (Mahoney, Varshney and Hind), Improving the Modeling Experience: Fairness Evaluation and Hyperparameter Tuning
• allreduce algorithm, Synchronous Allreduce
• API (application programming interface), Introduction to TensorFlow 2
  • (see also Horovod API)
  • choosing, Choosing the API
  • feature_column, Data Engineering
  • Keras, Keras API, Using the tf.keras.applications Module for Pretrained Models
  • model subclassing, Using the Imperative API-Defining a Model as a Class
  • RESTful (Representational State Transfer), Model Serialization
  • tf.keras, Keras API, Transfer Learning, Loading Example Data and Libraries
  • tf.saved_model, Serving Your TensorFlow Model
• asynchronous parameter server, Asynchronous Parameter Server

B

• batch size, Data Transformation and Input Processing
• batching datasets for training, Batching the Dataset for Training
• bias distributions, Visualizing the Learning Process Using TensorBoard
• build_model function, Model Serialization
• built-in training loop, Using the Built-In Training Loop

C

• __call__ function, Understanding Inheritance, Defining a Model as a Class
• callback objects
  • about, Callback Objects
  • EarlyStopping callback object, EarlyStopping
  • model checkpoint as, Model Checkpoint as a Callback
  • ModelCheckpoint class, ModelCheckpoint-ModelCheckpoint
• callbacks, Monitoring the Training Process
• child class, Understanding Inheritance
• Chollet, Francois, Keras API
• CIFAR-10 images, Loading the CIFAR-10 Images, ModelCheckpoint, TensorBoard, Visualizing Model Overfitting Using TensorBoard, Encapsulating ...