TensorFlow 1.5 Includes Mobile Version
Written by Kay Ewbank   
Tuesday, 30 January 2018

The latest release of TensorFlow has been released with TensorFlow Lite included, along with a new Eager Execution mode.  

TensorFlow is Google's open source tool that can be used for a wide range of parallel computations, including implementing neural networks and other AI learning methods. It is designed to make it easier to work with neural networks and is seen as more general and easier than other options.

The developers describe eager execution as an imperative, define-by-run interface where operations are executed immediately as they are called from Python. The benefits claimed for eager execution are that it makes debugging faster because you see run-time errors immediately, and can make use of Python tools for debugging. It also provides support for dynamic models using Python control flow. The other advantage it offers is support for custom and higher-order gradients. Writing about the new version, TensorFlow 1.5, on the Google Developers Blog, Laurence Moroney, Developer Advocate at Google said:

"With Eager Execution for TensorFlow enabled, you can execute TensorFlow operations immediately as they are called from Python. This makes it easier to get started with TensorFlow, and can make research and development more intuitive."

The other main improvement to this release is the inclusion of a built-in developer preview of TensorFlow Lite, TensorFlow's lightweight solution for mobile and embedded devices. Using TensorFlow Lite, you can take a trained TensorFlow model and convert it into a .tflite file which can then be executed on a mobile device with low-latency. This avoids the need for carrying out training on the lower powered device, and also avoids having to upload data to the cloud from the device.

Other improvements include built-in support for CUDA 9 and cuDNN 7 for those using GPU Acceleration on Windows or Linux; and enhancements to Accelerated Linear Algebra (XLA). 

 


tensorflow

More Information

TensorFlow Website

Related Articles

TensorFlow Incorporates Keras

TensorFlow Lite For Mobiles

TensorFlow Reaches Version 1

//No Comment - Should I use TensorFlow, AI Real Estate & Lip Reading 

 R Gets Notebooks & TensorFlow 

TensorFlow Course On Kadenze 

TPU Is Google's Seven Year Lead In AI 

TensorFlow 0.8 Can Use Distributed Computing 

TensorFlow - Googles Open Source AI And Computation Engine 

 

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.

 

Banner


GAN Theft Auto - The Neural Network Is The Game
16/07/2021

GAN Theft Auto is a fork of the Nvidia's GameGAN emulating dynamic game environments neural network but applied to Grand Theft Auto.



A Low Cost Hackable Watch
18/07/2021

Pine makes a range of community-supported hardware that is well worth knowing about, but its new PineTime SmartWatch offers a unique opportunity to program your wrist.


More News

 

square

 



 

Comments




or email your comment to: comments@i-programmer.info

Last Updated ( Wednesday, 19 September 2018 )