Release 18.06
The container image of NVIDIA Optimized Deep Learning Framework, powered by Apache MXNet, release 18.06, is available.
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
Note:
Container image
18.06-py2
contains Python 2.7;18.06-py3
contains Python 3.5. - NVIDIA CUDA 9.0.176 (see Errata section and 2.1) including CUDA® Basic Linear Algebra Subroutines library™ (cuBLAS) 9.0.333 (see section 2.3.1)
- NVIDIA CUDA® Deep Neural Network library™ (cuDNN) 7.1.4
- NCCL 2.2.13 (optimized for NVLink™ )
- ONNX exporter 0.1 for CNN classification models
Note:
The ONNX exporter is being continuously improved. You can try the latest changes by pulling from the main branch.
- Amazon Labs Sockeye sequence-to-sequence framework 1.18.22 (for machine translation)
- TensorRT 4.0.1
Driver Requirements
Release 18.06 is based on CUDA 9, which requires NVIDIA Driver release 384.xx.
Key Features and Enhancements
This Optimized Deep Learning Framework release includes the following key features and enhancements.
- NVIDIA Optimized Deep Learning Framework, powered by Apache MXNet container image version 18.06 is based on 1.2.0. Specifically, container image 18.06 has merged all commits on upstream Apache MXNet main, up to the creation point of the v1.2.0 branch, and all commits on that branch up to the 1.2.0 tag.
- Container includes TensorRT 4.0.1
- TensorRT integration examples for in-framework inference can be found in
/workspace/examples/tensorrt-integration
. This includes a LeNet-5 unit test and a ResNet-50 example. - Support added for DALI iterators.
- Ubuntu 16.04 with May 2018 updates
Announcements
Starting with the next major version of CUDA release, we will no longer provide updated Python 2 containers and will only update Python 3 containers.
Known Issues
Some of the unit tests available in /opt/mxnet/tests/python/{gpu,unittest}/*.py
require the SciPy Python library. For those that want to run the unit tests, first install the 1.0 version of SciPy by typing pip install scipy==1.0
.
The latest SciPy release, version 1.1, is not compatible with the unit tests.