Deep Learning AMI DLAMI Ubuntu 16 with Support by Supported Images
Product Overview
This is a repackaged open source software wherein additional charges apply for extended support with a 24 hour response time.
This image contains Deep Learning AMI (DLAMI) Ubuntu 16.04 and includes support.
This image is perfect for creating your deep learning models on a GPU instance such as: p2.xlarge, p2.8xlarge, p2.16xlarge, p3.2xlarge, p3.8xlarge, p3.16xlarge, g2.2xlarge, g2.8xlarge, g3.4xlarge, g3.8xlarge, g3.16xlarge, g3s.xlarge, g4dn.xlarge, g4dn.2xlarge, g4dn.4xlarge, g4dn.8xlarge, g4dn.12xlarge, g4dn.16xlarge and then changing to the inf1 instance types such as inf1.24xlarge, inf1.2xlarge, inf1.6xlarge, inf1.xlarge providing up to 40% lower cost per inference.
Deep Learning AMI has the folloframeworks (with CUDA 8.0, 9.0, 9.2, 10.0, 10.1 and cuDNN, NCCL):
Python 2 (python2)
Python 3 (python3)
Apache MXNet (Incubating) 1.6.0 with Gluon
TensorFlow 1.15 + Horovod
TensorFlow 2.0 + Horovod
PyTorch 1.3.1
Chainer 6.1.0
Keras 2.2.4.2
AWS Deep Learning AMI also comes with model debugging and hosting capabilities:
Apache MXNet (Incubating) Model Server 1.0
TensorFlow Serving 1.15
TensorBoard 1.15
GPU Drivers: NVIDIA Driver 418.87.02
Try our most popular AMIs on AWS EC2
- Ubuntu 22.04 AMI on AWS EC2
- Ubuntu 20.04 AMI on AWS EC2
- Ubuntu 18.04 AMI on AWS EC2
- CentOS 9 AMI on AWS EC2
- CentOS 8 AMI on AWS EC2
- CentOS 7 AMI on AWS EC2
- Debian 12 AMI on AWS EC2
- Debian 11 AMI on AWS EC2
- Debian 10 AMI on AWS EC2
- Debian 9 AMI on AWS EC2
- Red Hat Enterprise Linux 9 (RHEL 9) AMI on AWS EC2
- Red Hat Enterprise Linux 8 (RHEL 8) AMI on AWS EC2
- Red Hat Enterprise Linux 7 (RHEL 7) AMI on AWS EC2
- Oracle Linux 9 AMI on AWS EC2
- Oracle Linux 8 AMI on AWS EC2
- Oracle Linux 7 AMI on AWS EC2
- Amazon Linux 2023 AMI on AWS EC2
- Windows 2022 Server AMI on AWS EC2
- Windows 2019 Server AMI on AWS EC2
- Docker on Ubuntu 20 AMI on AWS EC2
- Docker on CentOS 7 AMI on AWS EC2
Version
Video
Operating System
Linux/Unix, Ubuntu 16.04 LTS
Delivery Methods