Overview
This course aims to teach learners about the transition from tf.Session to tf.function in TensorFlow. By the end of the course, students will understand the history of tf.Session, its drawbacks, and the motivation behind tf.function. They will also learn about the main classes and functions in the front-end used to generate graphs from tf.functions. The course covers topics such as TensorFlow Functions, Variables, Automatic Control, Calling Functions, and Variable Subclasses. The teaching method involves technical deep dives by TensorFlow Software Engineer Alexandre Passos. This course is intended for individuals interested in gaining a deeper understanding of TensorFlow and its functionalities.
Syllabus
Introduction
Session vs Runtime
Session Not Run
Partial Run
TensorFlow Functions
Variables
The compromise
The solution
The downsides
Walkthrough
Photograph
Capture
Phonegraph
Automatic Control
Calling Functions
Differentiating Functions
Concrete Functions
Code Base
Variable Lifting
Variable Subclasses
Variable Capture Scopes
Stateless Functions
Lift to Graph
Taught by
TensorFlow