v1.9

Table of Contents

  • Overview
  • Installation
  • QuickStart
  • Auto (Hyper-parameter) Tuning
    • Write Trial
    • Tuners
    • Assessors
    • Training Platform
    • Examples
    • WebUI
    • How to Debug
    • Advanced
      • Write a New Tuner
      • Write a New Assessor
      • Write a New Advisor
      • Write a New Training Service
      • Install Customized Algorithms as Builtin Tuners/Assessors/Advisors
      • How to install customized tuner as a builtin tuner
  • Neural Architecture Search
  • Model Compression
  • Feature Engineering
  • References
  • Use Cases and Solutions
  • FAQ
  • How to Contribute
  • ChangeLog
  • Release 1.9 - 10/22/2020
  • Release 1.8 - 8/27/2020
  • Release 1.7 - 7/8/2020
NNI
  • Docs »
  • Auto (Hyper-parameter) Tuning »
  • Advanced Features
  • Edit on GitHub

Advanced FeaturesΒΆ

  • Write a New Tuner
    • Customize Tuner
  • Write a New Assessor
  • Write a New Advisor
    • Example
  • Write a New Training Service
    • Overview
    • System architecture
    • Folder structure of code
    • Function annotation of TrainingService
    • TrialKeeper tool
    • Reference
  • Install Customized Algorithms as Builtin Tuners/Assessors/Advisors
    • Overview
    • Install customized algorithms as builtin tuners, assessors and advisors
    • 5. Use the installed builtin algorithms in experiment
    • Manage packages using nnictl package
  • How to install customized tuner as a builtin tuner
    • Prepare installation source and install package
    • Check the installed package
    • Use the installed tuner in experiment
Next Previous

© Copyright 2020, Microsoft Revision 98a72a1e.

Built with Sphinx using a theme provided by Read the Docs.