Skip to content

marlonsd/ActiveLearning

Repository files navigation

ActiveLearning

Implementation of common active learning strategies. The project is currently dedeveloped by the members of the Machine Learning Lab at IIT: https://ml.cs.iit.edu. Currently supported strategies include:

  1. Random sampling
  2. Uncertainty sampling
  3. Log-gain

Upcoming strategies:

  1. Query-by-committee
  2. Expected error reduction

Future work:

  1. Annotating features
  2. Tandem learning - annotating instances and features jointly

Current Limitations:

  1. The code currently assumes binary classification.

About

Active Learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published