NVIDIA, Facebook, soup up AI development with Caffe2

Source: NVIDIA blog post.

NVIDIA and Facebook have announced the fruits of joint work to advance artificial intelligence (AI) with Caffe2, a new AI deep learning framework contributed by Facebook to the open-source community.  The end result is that developers can train and iterate AI models faster than ever.

Caffe2 allows developers and researchers to create large-scale distributed training scenarios and build machine learning applications for edge devices to ensure that the AI can achieve what it is designed to do. The framework delivers near-linear scaling of deep learning training with 57x throughput acceleration on eight networked Facebook Big Basin AI servers with 64 NVIDIA Tesla P100 GPU accelerators.

"Providing AI-powered services on mobile is a complex data processing task that must happen within the blink of an eye. Increasingly, the processing of lightning-fast AI services requires GPU-accelerated computing, such as that offered by Facebook’s Big Basin servers, as well as highly optimised deep learning software that can leverage the full capability of the accelerated hardware," notes Kristin Bryson, NVIDIA's PR Director, in a blog post.

NVIDIA and Facebook have finetuned Caffe2 to take full advantage of the NVIDIA GPU deep learning platform. Caffe2 uses the latest NVIDIA Deep Learning SDK libraries — cuDNN, cuBLAS and NCCL — to deliver multi-GPU accelerated training and inference. "As a result, users can focus on developing AI-powered applications instead of trying to make Caffe2 deliver faster on their NVIDIA GPU systems," explained Bryson in the blog post.

Source: NVIDIA blog post. How fast Caffe2 works with NVIDIA's DGX-1 AI supercomputer.

As part of the companies’ collaboration, the NVIDIA DGX-1 AI supercomputer will be the first AI system to offer Caffe2 within the optimised software stack for deep learning. Caffe2 for DGX-1 will be available to customers via NVIDIA DGX-1 Container Registry.

NVIDIA will also be adding Caffe2 training through the NVIDIA Deep Learning Institute (DLI). "Through NVIDIA’s Deep Learning Institute, we’ve helped more than 10,000 developers worldwide learn to use frameworks to design, train and deploy neural network-powered machine learning for a variety intelligent applications and services.

"Starting at our GPU Technology Conference on May 8 to 11, we’re adding Caffe2 training to the DLI curriculum. Members of our DLI team will be on hand to help developers get their hands dirty with Caffe2 through a self-paced lab," said Bryson.