Skip to content

Máking changes in README.md file #2014

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
loccbfx78906-glitch wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Máking changes in README.md file #2014

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 10 additions & 4 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,5 @@
# Making changes in README.md file

# What's new in TensorFlow 2.x

The following are all the changes coming in TensorFlow 2.x. Let's have a closer look at them:
Expand All @@ -14,14 +16,16 @@ The following are all the changes coming in TensorFlow 2.x. Let's have a closer
* Tidying up the API
* The conversion tool
* Alternative variable scoping




## Eager Execution



Lack of eager execution was one of the main complaints against TensorFlow. We all can relate. Having to execute the whole graph and then trying to debug based on the errors was very tedious. Especially, since values of intermediate results haven't been accessible without printing them out by mixing in debug statements into the production code.

With TensorFlow 2.0, eager execution is activated by default and the very cool thing is that the code nearly doesn't change. Under the hood, you are just working with so-called "EagerTensors" instead of "Tensors" but since they share the same interface, the difference is barely noticeable. Even in execution speed, the difference is hard to see.
With TensorFlow 2.0, eager execution is activated by default and the very cool thing is that the code nearly doesn't change. Under the hood, you are just working with so-called "EagerTensors" instead of "Tensors" but since they share the same interface, the difference is barely noticeable. Even in execution speed, the difference is hard to see.

This means, from now on, TensorFlow code can be used and debugged as ordinary python code (using numpy for example). This is one aspect of making TensorFlow more pythonic.

Expand All @@ -32,7 +36,8 @@ Below there are two tasks. I highly recommend doing them because while watching
1. [Watch me coding](https://www.youtube.com/watch?v=J3_b4461qxU)


2. [Code yourself](https://github.com/romeokienzler/TensorFlow/blob/master/notebooks/tf2.eagerexec.ipynb)

1. [Code yourself](https://github.com/romeokienzler/TensorFlow/blob/master/notebooks/tf2.eagerexec.ipynb)

## Integration of the Keras API

Expand All @@ -49,7 +54,8 @@ For now, just follow along the video and code exercise below to get an idea how
1. [Watch me coding](https://www.youtube.com/watch?v=D4mJZQdgV0Y)


2. [Code yourself](https://github.com/romeokienzler/TensorFlow/blob/master/notebooks/tf2.keras.ipynb)

1. [Code yourself](https://github.com/romeokienzler/TensorFlow/blob/master/notebooks/tf2.keras.ipynb)



Expand Down