5. Catalit LLC
KEY POINTS
• Public 2.0 design process
• Eager execution
• Remove deprecated APIs & reduce the amount of duplication
• Compatibility and continuity withTensorflow 1.x
• Compatibility with 1.x exported models
• No more tf.contrib
• More platforms and languages
8. Catalit LLC
TODAY
• Public 2.0 design process
• Eager execution
• Remove deprecated APIs & reduce the amount of duplication
• Compatibility and continuity withTensorflow 1.x
• Compatibility with 1.x exported models
• No more tf.contrib
• More platforms and languages
9. Catalit LLC
KEY POINTS
• Public 2.0 design process
• Eager execution
• Remove deprecated APIs & reduce the amount of duplication
• Compatibility and continuity withTensorflow 1.x
• Compatibility with 1.x exported models
• No more tf.contrib
• More platforms and languages
10. Catalit LLC
PUBLIC 2.0 DESIGN PROCESS
https://groups.google.com/a/tensorflow.org/forum/#!forum/discuss
https://github.com/tensorflow/community/blob/master/governance/TF-RFCs.md
https://github.com/tensorflow/community/tree/master/rfcs
11. Catalit LLC
WHAT IT MEANS FORYOU
• Final release expected in summer 2019
• If you’re aTF 1.x developer
=> start updating your code as soon as RFPs are accepted
• There is a conversion tool from 1.x to 2.0 (won’t be
perfect, but hey…)
13. Catalit LLC
KEY POINTS
• Public 2.0 design process
• Eager execution
• Remove deprecated APIs & reduce the amount of duplication
• Compatibility and continuity withTensorflow 1.x
• Compatibility with 1.x exported models
• No more tf.contrib
• More platforms and languages
14. Catalit LLC
EAGER EXECUTION
• Available since late 2017
• Following Pytorch and Chainer
• Imperative
• Define-by-run
• No static graph & session
15. Catalit LLC
EAGER EXECUTION
• Faster debugging with Python tools
• Dynamic models with Python control flow
• Support for custom and higher-order gradients
16. Catalit LLC
IS GRAPH GOING AWAY?
• No, you can still define models using the traditional
static graph
• tf.compat.v1.disable_eager_execution()
21. Catalit LLC
WHAT IT MEANS FORYOU
• Debug like Numpy, scale likeTensorflow
• Easier to build and test custom models
• Slower than static graph mode
22. Catalit LLC
KEY POINTS
• Public 2.0 design process
• Eager execution
• Remove deprecated APIs & reduce the amount of duplication
• Compatibility and continuity withTensorflow 1.x
• Compatibility with 1.x exported models
• No more tf.contrib
• More platforms and languages
27. Catalit LLC
WHAT IS KERAS?
https://towardsdatascience.com/deep-learning-framework-power-scores-2018-23607ddf297a
28. Catalit LLC
WHAT IS KERAS
• Keras is an API specification to design deep learning model
• https://keras.io/
=> independent reference implementation (usingTF or CNTK as backend)
• tf.keras
=>Tensorflow implementation of the same API spec
• Other frameworks implement some version of the API
34. Catalit LLC
DIFFERENCES
• Support for Eager Execution
• tf.dataVS python data generators
• Model Exporting
• Compatible with Feature Columns
• Compatible with Estimators
35. Catalit LLC
KEY POINTS
• Public 2.0 design process
• Eager execution
• Remove deprecated APIs & reduce the amount of duplication
• Compatibility and continuity with Tensorflow 1.x
• Compatibility with 1.x exported models
• No more tf.contrib
• More platforms and languages
https://medium.com/tensorflow/whats-coming-in-tensorflow-2-0-d3663832e9b8