Swift for TensorFlow is now Open Sourced on GitHub
Overview
- Swift for TensorFlow, demo’d at the TensorFlow conference last month, has been open sourced on GitHub
- It’s still in it’s very early stages so developing entire ML frameworks is beyond it’s scope currently
- Watch the video below to get an introduction and feel for this release
Introduction
Swift is an open source programming language which has really taken off in the last couple of years. It has a large, and ever expanding user base. And TensorFlow, as you will no doubt be aware, is one of the most popular open source libraries used in machine learning. So combining the two together was a no-brainer for the folks at TensorFlow.
To learn about Swift and how it is used in Data Science in a thorough and structured manner, you can enrol in this free course: Learn Swift for Data Science
Swift for TensorFlow was demo’d at the TensorFlow Conference last month and the team behind the technology has now open sourced the code on GitHub for the entire community. Their aim is to provide a new interface to TensorFlow that will build on it’s already awesome capabilities, while taking it’s usability to a whole new level.
According to the official blog post by the TensorFlow team, “Swift for TensorFlow provides a new programming model that combines the performance of graphs with the flexibility and expressivity of Eager execution, with a strong focus on improved usability at every level of the stack”. Note that this isn’t just a TensorFlow API wrapper written in the Swift language . The team has added compiler and language enhancements to Swift with the aim of providing a top notch user experience for data scientists and machine learning developers.
You can access the GitHub repository here and watch the TensorFlow conference launch in the below video:
Our take on this
This is still in it’s very nascent stages so isn’t yet ready to be written into deep learning models. The team admits that the goals it has in mind while launching this are still a while away from being achieved. But there is a lot of potential here that is as yet untapped.
What I liked about this release is that the team has documented each step in extreme detail with the assumption that most of the users will not be familiar with Swift, or wouldn’t have used it before.
Subscribe to AVBytes here to get regular data science, machine learning and AI updates in your inbox!
Senior Editor at Analytics Vidhya. Data visualization practitioner who loves reading and delving deeper into the data science and machine learning arts. Always looking for new ways to improve processes using ML and AI.
