At its F8 developer conference in San Jose today, Facebook is announcing the launch of Caffe2, a new open source framework for deep learning, a trendy type of artificial intelligence (AI). Deep learning generally involves training artificial neural networks on lots of data, like photos, and then getting them to make inferences about new data.
Today’s announcement builds on Facebook’s contributions to the Torch open source deep learning framework and more recently the PyTorch framework that the Facebook Artificial Intelligence Research (FAIR) group conceived. And last year Facebook talked about a system called Caffe2go.
But Caffe2 is a bit different from PyTorch.
“PyTorch is great for research, experimentation and trying out exotic neural networks, while Caffe2 is headed towards supporting more industrial-strength applications with a heavy focus on mobile,” Facebook AI Platform engineering lead Yangqing Jia wrote in a comment on Hacker News. “This is not to say that PyTorch doesn’t do mobile or doesn’t scale or that you can’t use Caffe2 with some awesome new paradigm of neural network, we’re just highlighting some of the current characteristics and directions for these two projects. We plan to have plenty of interoperability and methods of converting back and forth so you can experience the best of both worlds.”
There are several other open source deep learning frameworks for people to use for all kinds of purposes. In 2015 Google made a splash when it open-sourced its own framework, TensorFlow.
Earlier today Facebook talked about how it’s using AI for augmented reality (AR), among other things. But Facebook is also using AI in existing products, including speech recognition in Oculus and image recognition on Facebook.
Public cloud infrastructure provider Amazon Web Services (AWS) has added experimental support for Caffe2 in its Deep Learning AMI (Amazon machine image). Microsoft Azure, another public cloud, has added Caffe2 support to its Data Science Virtual Machine (DSVM).
Nvidia, whose graphics processing units (GPUs) can go inside Facebook’s servers for deep learning, will be helping people learn about Caffe2, Nvidia accelerated computing general manager and vice president Ian Buck wrote in a blog post. Qualcomm said it’s working with Facebook to optimize Caffe2 and Qualcomm’s Snapdragon neural processing engine (NPE) framework, and Intel said it will integrate its Math Kernel Library (MKL) into Caffe2.