Tech giant Google says it will release a machine learning system called TensorFlow into the open source platform to allow researchers and developers to freely use the source code to further develop and upgrade the system.

"We use TensorFlow for everything from speech recognition in the Google app, to Smart Reply in inbox, to search in Google Photos. It allows us to build and train neural nets up to five times faster than our first-generation system, so we can use it to improve our products much more quickly," wrote Google CEO Sundar Pichai on the company's official blog.  

Like Us on Facebook

TensorFlow is the successor of the DistBelief system that was used by Google to sniff out spam on emails, recognize videos on YouTube, and improve voice recognition.

However, DistBelief is closely integrated into Google's internal systems and is quite difficult to configure. With TensorFlow, Google is releasing the same version of DistBelief under the Apache 2.0 open source license in order to encourage those who are interested to help the company develop the system.

The TensorFlow system can be applied to any dataset that can be reduced to a simple flowchart and comes within the C++ or Python framework. Since it is released on the open source platform, almost anyone can use it as long as they adhere to the Apache 2.0 regulations.

Google claims that the software can run on almost any platform ranging from smartphones, single CPU systems, or even data center clusters.

But why is Google giving out its codes for free?

Releasing a software on the open source platform means that it can reach a broader audience. From that audience, Google can pick those who have potential and add it into the company's workforce. This is just one of the many incentives for releasing a software in an open source environment. 

TagsMachine Learning, open source, TensorFlow