Jump to Content
DeciBack to Console
DocumentationAPI Reference
Back to ConsoleLog InDeci
Log In
DocumentationAPI Reference

Deci Platform Quickstart Guide

  • Quickstart
    • Step 1 – Launching the Deci Platform
    • Step 2 – Exploring an Example Project
    • Step 3 – Adding Your Model
    • Step 4 – Optimizing Your Model
    • Step 5 – Benchmarking Your Model(s)
  • Model Zoo – Practicing with Pre-trained Models
    • Model Zoo Models
  • Lab – Managing and Optimizing Your Models
  • Insights Tab – Benchmarking Your Models
    • Overview Section
    • Performance Section
    • Accuracy Metrics Section
    • Finding the Optimal Hardware Benchmark

Infery - A python inference engine

  • Introducing Infery
  • Installing Infery
  • Quick Start
  • Loading Model
  • Error Handling
  • Infery Examples On GitHub

Release Notes

  • Deci Platform Release Notes
  • Supported Deep Learning Frameworks

FAQs

  • What is?
    • What is the Deci Platform?
    • What is Model Runtime Performance?
    • What is AutoNAC?
    • What is Infery?
  • How Does Deci Work?
    • Does Deci accelerate the training of deep learning models?
    • Which model checkpoint does the Deci platform support?
    • Efficient Frontier Methodology
  • How Do I?
    • How can I compare a model across various hardware?
    • How Do I Upload My PyTorch Model into Deci?

Troubleshooting

  • I can’t log-in
  • My model upload fails
  • My model fails to benchmark
  • My model benchmark stats are empty
  • I can’t select a specific hardware or batch size to benchmark in the Insights screen.
  • My model optimization fails

Step 5 – Benchmarking Your Model(s)

Suggest Edits

Now that you have an optimized Deci version of your model, you can explore its performance metrics.
You may refer to Viewing Benchmark Insights and Exploring an Example Project for description of an optimized Deci model’s performance metrics/benchmarks.

Updated over 1 year ago