Facebook PyTorch 1.0 – new open-source AI framework for developers

Facebook PyTorch 1.0 – new open-source AI framework for developers

On May 2nd, 2018, at Facebook’s F8 developer conference, the company announced PyTorch 1.0, a new open-source AI framework for developers. Facebook has invested a lot in the development of machine learning and artificial intelligence technology and is now rolling out its newest product in order to share the results of this research with the public. What is it and what does it do?

The core of PyTorch 1.0

 

The Python-based PyTorch 1.0 combines training, reasearch and production in one framework. This eliminates the complicated process of moving from one framework to another. It provides the developers a base for building different types of AI projects. The technology behind it integrates an open source machine learning library with a deep learning framework.

What does it actually do?

 

Facebook already utilizes the technology behind PyTorch 1.0. Currently, it provides six billion translations of posts on the platform daily. PyTorch 1.0 has also been trained in imaged recognition for example. The framework will provide developers with a wide variety of tools. These will help them optimize their performance as well as do experimenting. These tools include things like data sets and code libraries.

PyTorch 1.0 could very well be an important tool in aiding the developers harness the power of machine learning and AI. Microsoft’s Azure for example will support the framework. The first version of PyTorch was introduced last year and now the beta version of PyTorch 1.0 will roll out in the next couple of months.

To read more about the release of PyTorch 1.0 and its properties, you can click here, here or here for more articles on the subject. If you are interested in harnessing the power of AI for your field service business but lack the know-how, check out how Husky can help you do that.

Want to see Husky in Action? Contact today for a personalized demo.