Release 17.05

The container image of NVIDIA Optimized Deep Learning Framework, powered by Apache MXNet, release 17.05, is available.

NVIDIA Optimized Deep Learning Framework, powered by Apache MXNet container image version 17.05 is based on Apache MXNet 0.9.3a+.

Contents of the Optimized Deep Learning Framework container

This container image contains the complete source of the version of NVIDIA Optimized Deep Learning Framework, powered by Apache MXNet in /opt/mxnet. It is pre-built and installed to the Python path. The container also includes the following:

• Ubuntu 16.04
• NVIDIA CUDA^® 8.0.61
• NVIDIA CUDA^® Deep Neural Network library™ (cuDNN) 6.0.21
• NCCL 1.6.1 (optimized for NVLink™ )

Key Features and Enhancements

This NVIDIA Optimized Deep Learning Framework, powered by Apache MXNet release includes the following key features and enhancements.

• Latest cuDNN release
• Improved IO pipeline for increased multi-GPU performance
• Optimized SGD weight update
• Added the nccl_allreduce option for gradient communication
• Added support for dilated deconvolutions
• Improved convolutional neural network (CNN) performance by removing unnecessary computations
• Added options to show the cuDNN algorithms that are chosen for convolutions
• Ubuntu 16.04 with April 2017 updates

Known Issues

There are no known issues in this release.