CentML DeepView

CentML DeepView provides an integrated experience which allows ML practioners to:

• Visually identify model bottlenecks

• Perform rapid iterative profiling

• Understand energy consumption and environmental impacts of training jobs

• Predict deployment time and cost to cloud hardware

Getting Started

Follow the instructions depending on your setup

• Local workstation GPU

• Remote GPU with SSH access

• Clusters, containers, and other setups where SSH is not possible

DeepView.Profile

Deepview.Profile is our free and open source tool provides easier way to identify bottlenecks and perform rapid iterative profiling for Deep Learning. To find out more, visit DeepView.Profile

DeepView.Predict

DeepView.Predict is a tool that predicts a deep neural network’s training iteration execution time on a given GPU. It currently supports PyTorch. To find out more, visit DeepView.Predict

Index

Getting Started

• Quick Start
  • Installation
  • Verifying Installation

Profiler How-to Guide

• Providers
• Runtime Report
• Memory Report

References

• Python API
• Index