This item is a repackaged software; additional charges apply for support. AWS Deep Learning AMIs are the industry leader in application packaging. They are built and optimized for building, training, debugging, and serving deep learning models in EC2 with popular frameworks such as TensorFlow, MXNet, PyTorch, and more. Deep learning frameworks are installed in Conda environments to provide a reliable and isolated environment for practitioners. The AWS Deep Learning AMI is provided at no additional charge to Amazon EC2 users. Deep Learning frameworks are with the latest versions of NVIDIA CUDA, cuDNN, and Intel acceleration libraries such as MKL-DNN for high performance across CPU and GPU AWS EC2 instance types.
Below are the core components of AWS Deep Learning AMI:
Popular deep learning frameworks including TensorFlow (1.x, 2.x), PyTorch (1.x), and MXNet (1.x) performance tuned for use in AWS infrastructures. They are organized into Conda environments that are configured to be used out-of-the-box.
Built-in support for AWS Inferentia with the Inf1 instance family.
AWS Deep Learning Tools including AWS Elastic Fabric Adapter (EFA) and AWS Neuron.
NVIDIA Deep Learning Software including NVIDIA GPU Driver, CUDA Toolkit, cuDNN, NCCL, and Fabric Manager.”