New in AWS Deep Learning AMIs: TensorFlow 1.10, PyTorch with CUDA 9.2, and More
The AWS Deep Learning AMIs for Ubuntu and Amazon Linux now come with newer versions of the following deep learning frameworks and interfaces: TensorFlow 1.10 optimized for AWS for higher performance, Horovod 0.13.11 with OpenMPI 3.1.0 optimized for distributed multi-GPU TensorFlow training on Amazon EC2 P3 instances, PyTorch with CUDA 9.2 optimized for model training on Amazon EC2 P3 instances, Chainer 4.3.1, and Keras 2.2.2.
The AWS Deep Learning AMIs for Ubuntu and Amazon Linux now come with newer versions of the following deep learning frameworks and interfaces: TensorFlow 1.10 optimized for AWS for higher performance, Horovod 0.13.11 with OpenMPI 3.1.0 optimized for distributed multi-GPU TensorFlow training on Amazon EC2 P3 instances, PyTorch with CUDA 9.2 optimized for model training on Amazon EC2 P3 instances, Chainer 4.3.1, and Keras 2.2.2.
Faster training with optimized TensorFlow 1.10
The Deep Learning AMIs come with an optimized build of TensorFlow 1.10, custom built to accelerate deep learning applications on Amazon EC2 C5 and P3 instances. Deep Learning AMIs automatically deploy the TensorFlow build optimized for the EC2 instance of your choice when you activate the TensorFlow virtual environment for the first time.
For developers looking to scale their TensorFlow training from a single GPU to multiple GPUs, the AWS Deep Learning AMIs come with Horovod, optimized for distributed training using Amazon EC2 P3 instances. You can read more about our custom TensorFlow optimizations for AWS in this blog post.
Latest in framework updates
Deep Learning AMIs now support the latest PyTorch 0.4.1 pre-configured with NVidia CUDA 9.2, cuDNN 7.1.4, and NCCL 2.2.13 for accelerated deep learning on Amazon EC2 P3 instances. Also Chainer is now upgraded to version 4.3.1, optimized for high performance across Amazon EC2 instance families. You can read more about the Chainer optimizations on Deep Learning AMIs in this blog post.
AWS Deep Learning AMIs also support Apache MXNet 1.2.1 with Gluon, Microsoft Cognitive Toolkit (CNTK) 2.5.1, Caffe 1.0, Caffe2 0.8.1 and Theano 1.0.1 —all pre-installed and fully-configured for you to start developing your deep learning models in minutes while taking full advantage of the computational power of Amazon EC2 instances.
Get started quickly with the AWS Deep Learning AMIs using the getting-started guides and beginner-to-advanced level tutorials in our developer guide. You can also subscribe to our discussion forum to get launch announcements and post your questions.