Blog Post: http://kuldeepsinghsidhu.blogspot.com
Google Colab: Jupyter Notebook
Eager Execution (EE) enables you to run operations immediately. In TensorFlow, you have to create a graph and run it within a session in order to execute the operations of the graph. On the other hand, EE enables you to run operations directly and inspect the output as the operations are executed. This is very useful especially for debugging. Moreover, EE is pythonic and intergrates pretty well with numpy so it makes programming really easy and flexible. The next version of TenosrFlow "2.0" will enable EE by default. From Google AI article, these are some benefits of using EE
- Fast debugging with immediate run-time errors and integration with python tools
- Support for dynamic models using easy-to-use Python control flow
- Strong support for custom and higher-order gradients
- Almost all of the available TensorFlow operations
Kuldeep Singh Sidhu
Github: github/singhsidhukuldeep
https://github.com/singhsidhukuldeep
Website: Kuldeep Singh Sidhu (Website)
http://kuldeepsinghsidhu.com
LinkedIn: Kuldeep Singh Sidhu (LinkedIn)
https://www.linkedin.com/in/singhsidhukuldeep/