Microsoft Cognitive Toolkit Release 17.03

The NVIDIA container image of Microsoft Cognitive Toolkit, release 17.03, is available.

Microsoft Cognitive Toolkit container image version 17.03 is based on Microsoft Cognitive Toolkit 2.0.beta12.0.

Contents of Microsoft Cognitive Toolkit

This container image contains the source files that are used to build the Microsoft Cognitive Toolkit in the /opt/cntk directory. The Microsoft Cognitive Toolkit binaries are pre-built and installed into the /usr/local/cntk directory.

A Python 3.4 Conda environment (including the Microsoft Cognitive Toolkit module) is installed in the /opt/conda/envs/cntk-py34 directory. This is included in PATH, so by default the python command executes a Python 3.4 interpreter. To use the base Ubuntu Python 2.7 binary, you can either adjust PATH or explicitly call /usr/bin/python. This image of the Microsoft Cognitive Toolkit, 2.0beta12.0, is built with the following options enabled:

1-bit SGD
Async SGD
Python 3 API
OpenCV
Kaldi

The container also includes the following:

Ubuntu 16.04
NVIDIA CUDA^® 8.0.61
NVIDIA CUDA^® Deep Neural Network library™ (cuDNN) 6.0.20
NCCL 1.6.1 (optimized for NVLink™ )

Key Features and Enhancements

This Microsoft Cognitive Toolkit release includes the following key features and enhancements.

Dilated convolution support (carried over from 17.02)
GPU memory consumption optimizations
NCCL integration for MASGD Solver
Fixes in provided example and tutorial scripts
Ubuntu 16.04 with February 2017 updates

Known Issues

There is a known issue in 17.03 that causes corruption in networks incorporating residual connections, such as ResNet. As a result, they fail to train to full accuracy.