Contents Menu Expand Light mode Dark mode Auto light/dark mode
datagnosis 0.0.3 documentation
datagnosis 0.0.3 documentation
  • Tutorials
    • Getting started with tabular data
    • Image data
    • A real tabular data example
  • DataHandlers
    • DataHandler
  • Hardness Characterization Methods
    • Area Under the Margin (AUM)
    • Conf Agree
    • Confident Learning
    • Data IQ
    • Data Maps
    • Error L2-Norm (EL2N)
    • Forgetting
    • Gradient Normed (GraNd)
    • Large Loss
    • Prototypicality
    • Variance of Gradients (VOG)
    • Active Learning Guided by Local Sensitivity and Hardness (ALLSH)
Back to top
Edit this page

Hardness Characterization Methods#

General purpose#

  • Area Under the Margin (AUM)
  • Conf Agree
  • Confident Learning
  • Data IQ
  • Data Maps
  • Error L2-Norm (EL2N)
  • Forgetting
  • Gradient Normed (GraNd)
  • Large Loss
  • Prototypicality
  • Variance of Gradients (VOG)

Image specific#

  • Active Learning Guided by Local Sensitivity and Hardness (ALLSH)
Next
datagnosis.plugins.generic.plugin_aum module
Previous
datagnosis.plugins.core.datahandler module
Copyright © 2023, vanderschaar-lab
Made with Sphinx and @pradyunsg's Furo
On this page
  • Hardness Characterization Methods
    • General purpose
    • Image specific