Get started with model compilation with Neo

Apache MXNet

PyTorch

• Deploying pre-trained PyTorch vision models with Amazon SageMaker Neo

Inf1 Instance

• Deploying pre-trained PyTorch vision models with Amazon SageMaker Neo On Inf1 Instance
• Compile Model with Default Settings
• Compile Model for Multiple Cores using Compiler Options
• Compiling HuggingFace models for AWS Inferentia with SageMaker Neo
• Conclusions
• Compile and Deploy a MXNet model on Inf1 instances
• Construct a script for training and hosting
• Deploy the trained model on Inf1 instance for real-time inferences

TensorFlow

• Using SageMaker Neo to Compile a Tensorflow U-Net Model