This repository contains a series of articles as well as supporting materials (source code, shell scripts, and data files) describing various aspects of deployment of deep learning applications on Genesis Cloud GPU-enabled infrastructure.
The following articles are currently available:
Article 1 Installation and basic use of CUDA, PyTorch, and torchvision
Article 2 Deployment techniques for PyTorch models using TorchScript
Article 3 Deployment techniques for PyTorch models using TensorRT
Article 4 Using Triton for production deployment of TensorRT models
Article 5 Introduction to transformer models and Hugging Face library
Article 6 Using ONNX Runtime for deployment and optimization of transformer models
The repository is work in progress.