Google open-sourced its TensorFlow machine studying framework again in 2015 and it rapidly grew to become one of the well-liked platforms of its sort. Enterprises that wished to make use of it, nonetheless, needed to both work with third events or do it themselves. To assist these firms — and seize a few of this profitable market itself — Google is launching TensorFlow Enterprise, which incorporates hands-on, enterprise-grade assist and optimized managed companies on Google Cloud.
One of the crucial vital options of TensorFlow Enterprise is that it’s going to provide long-term assist. For some variations of the framework, Google will provide patches for as much as three years. For what seems to be an extra charge, Google will even provide engineering help from its Google Cloud and TensorFlow groups to firms which can be constructing AI fashions.
All of this, after all, is deeply built-in with Google’s personal cloud companies. “As a result of Google created and open-sourced TensorFlow, Google Cloud is uniquely positioned to supply assist and insights immediately from the TensorFlow workforce itself,” the corporate writes in at present’s announcement. “Mixed with our deep experience in AI and machine studying, this makes TensorFlow Enterprise the easiest way to run TensorFlow.”
Google additionally consists of Deep Studying VMs and Deep Studying Containers to make getting began with TensorFlow simpler and the corporate has optimized the enterprise model for Nvidia GPUs and Google’s personal Cloud TPUs.
At present’s launch is one more instance of Google Cloud’s deal with enterprises, a transfer the corporate accelerated when it employed Thomas Kurian to run the Cloud companies. After years of largely ignoring the enterprise, the corporate is now clearly taking a look at what enterprises are battling and the way it can adapt its merchandise for them.