In this webinar series, we will do an in-depth presentation of Triton and its main components. We also will demonstrate how to quick-start and use Triton on real word applications on-prem, on the cloud, or in a mixed environment. In addition, we provide you with the scripts and code to jumpstart your Triton expertise.

TensorRT is an optimizing compiler of neural networks for inference. NVIDIA TensorRT-based applications perform up to 36X faster than CPU-only platforms during inference, enabling you to optimize neural network models trained on all major frameworks, calibrate for lower precision with high accuracy, and deploy to hyperscale data centers, embedded platforms, or automotive product platforms.

The NVIDIA Triton™ Inference Server is an open-source, high-performance inference-serving software that helps standardize model deployment and execution and delivers fast and scalable AI in production. Triton supports any ML/DL models, frameworks, and hardware (CPU, GPU, servers, cloud, edge, embedded) and can work with any middleware (bare metal, virtualization, containers, microservices).

• Overview of TensorRT and its integration with Pytorch and TensorFlow

• Rapid Prototyping and Testing of ML Models With PyTriton

• A complete, reproducible demo of how one can take a Pytorch model, compile with TensorRT and inference in production in Triton.

• Review and Real Enterprise Customer Cases

Content

content goes here