Updated AWS Deep Learning AMIs With TensorFlow 1.5 and Microsoft Cognitive Toolkit 2.4
The AWS Deep Learning AMIs now come with TensorFlow 1.5, and Microsoft Cognitive Toolkit 2.4, with support for NVIDIA CUDA 9 and cuDNN 7 drivers. AMI users can now take advantage of the mixed precision training on the V100 Volta GPUs powering EC2 P3 instances. In our earlier tests for TensorFlow 1.5 on Volta, training the ResNet-50 benchmark with synthetic ImageNet data in FP16 mode on a p3.8xlarge instance was 1.8x times faster than training with TensorFlow 1.4.1.
The AWS Deep Learning AMIs now come with TensorFlow 1.5, and Microsoft Cognitive Toolkit 2.4, with support for NVIDIA CUDA 9 and cuDNN 7 drivers. AMI users can now take advantage of the mixed precision training on the V100 Volta GPUs powering EC2 P3 instances. In our earlier tests for TensorFlow 1.5 on Volta, training the ResNet-50 benchmark with synthetic ImageNet data in FP16 mode on a p3.8xlarge instance was 1.8x times faster than training with TensorFlow 1.4.1.
The AWS Deep Learning AMIs are designed to help you build stable, secure and scalable deep learning applications. Visit our documentation guide to select the right AMI for your project and get started with deep learning on AWS.