Model Analyzer Documentation

Installation	Getting Started	User Guide	Additional Resources

Installation

See the Installation Guide for details on how to install Model Analyzer.

Getting Started

• The Quick Start Guide will show you how to use Model Analyzer to profile, analyze and report on a simple PyTorch model.

• Watch the Optimizing Triton Deployments video for a step-by-step guide of how the optimal configuration for a BERT model can be found.

User Guide

The User Guide describes how to configure Model Analyzer, choose launch and search modes, and describes what metrics are captured as well as the different reports that can be generated.

• Model Analyzer CLI

• Launch Modes

• Configuring Model Analyzer

• Model Analyzer Metrics

• Model Config Search

• Checkpointing

• Model Analyzer Reports

• Deployment with Kubernetes

Resources

The following resources are recommended:

• Perf Analyzer: Perf Analyzer is a CLI application built to generate inference requests and measures the latency of those requests and throughput of the model being served.