Caffe2 Overview

The NVIDIA Deep Learning SDK accelerates widely-used deep learning frameworks such as Caffe2™ .

Caffe2 is a deep-learning framework designed to easily express all model types, for example, convolutional neural networks (CNNs), recurrent neural networks (RNNs), and more, in a friendly Python-based API, and execute them using a highly efficiently C++ and CUDA® backend.

Caffe2 includes a flexible API that lets users define models for inference or training using expressive, high-level operations. The Python interface allows easy control and visualization of the inference or training process.

Caffe2 supports single and multi-GPU execution, along with support for multi-node execution.

This document describes the key features, software enhancements and improvements, any known issues, and how to run this container.