New AWS Deep Learning AMIs: Amazon Linux 2, TensorFlow 1.13.1, MXNet 1.4.0, and Chainer 5.3.0
The AWS Deep Learning AMIs are now available on Amazon Linux 2, the next generation of Amazon Linux, in addition to Amazon Linux and Ubuntu. In addition, the AWS Deep Learning AMIs now come with MXNet 1.4.0, Chainer 5.3.0, PyTorch 1.0.1, and TensorFlow 1.13.1, which is custom-built directly from source and tuned for high-performance training across Amazon EC2 instances.
The AWS Deep Learning AMIs are now available on Amazon Linux 2, the next generation of Amazon Linux, in addition to Amazon Linux and Ubuntu. In addition, the AWS Deep Learning AMIs now come with MXNet 1.4.0, Chainer 5.3.0, PyTorch 1.0.1, and TensorFlow 1.13.1, which is custom-built directly from source and tuned for high-performance training across Amazon EC2 instances.
On CPU instances, TensorFlow 1.13 is custom-built directly from source to accelerate performance on Intel Xeon Platinum processors that power EC2 C5 instances. Training a ResNet-50 model with synthetic ImageNet data using the Deep Learning AMI results in 9.4X faster throughput than stock TensorFlow 1.13 binaries. GPU instances come with an optimized build of TensorFlow 1.13 that is configured with NVIDIA CUDA 10 and cuDNN 7.4 to take advantage of mixed precision training on Volta V100 GPUs powering EC2 P3 instances. For developers looking to scale their TensorFlow training to multiple GPUs, the Deep Learning AMIs come with the Horovod distributed training framework. The framework is fully optimized to efficiently use distributed training cluster topologies composed of Amazon EC2 P3 instances. Training a ResNet-50 model using TensorFlow 1.13 and Horovod in the Deep Learning AMI results in 27% faster throughput than stock TensorFlow 1.13 on 8 nodes.
AWS Deep Learning AMIs now come with the latest release of Apache MXNet 1.4 that bring improvements to performance and ease-of-use. MXNet 1.4 adds Java bindings for inference, Julia bindings, experimental control flow operators, JVM memory management, and many more under-the-hood enhancements. This release also improves MXNet support for Intel MKL-DNN with improved graph optimization and quantization. This feature reduces memory usage and improves inference time without a significant loss in accuracy.
Get started quickly with the AWS Deep Learning AMIs using the getting-started guides and beginner to advanced level tutorials in our developer guide. When you activate a Conda environment, the Deep Learning AMIs automatically deploy higher-performance builds of frameworks, optimized for the EC2 instance of your choice. For a complete list of frameworks and versions supported by the AWS Deep Learning AMI, see the release notes. You can also subscribe to our discussion forum to get launch announcements and post your questions.