Welcome to PyRelationAL’s documentation!

PyRelationAL is a python active learning library developed by Relation Therapeutics for rapidly implementing active learning pipelines from data management, model development (and Bayesian approximation), to creating novel active learning strategies.

Notes

• What is Active Learning?
• Installation
• Quickstart and introduction by example
• Using your own datasets with PyRelationAL
• Defining learning models compatible with PyRelationAL
• Creating your own active learning strategies with PyRelationAL
• Benchmark datasets and AL task configurations

Package modules

• pyrelational.data_managers
  • Data Manager
• pyrelational.datasets
  • Classification datasets
  • Regression datasets
  • Benchmark DataManager
• pyrelational.model_managers
  • Abstract Model Manager
  • Pytorch Lightning Model
  • Ensemble Models
  • MCDropout Models
• pyrelational.informativeness
  • Informativeness functions for regression tasks
  • Informativeness functions for classification tasks
  • Task agnostic informativeness functions
• pyrelational.strategies
  • Abstract strategy
  • Strategies for regression tasks
  • Strategies for classification tasks
  • Task-agnostic strategies
• pyrelational.oracles
  • Abstract Oracle
  • Benchmark Oracle
• pyrelational.pipeline
  • Pipeline

Indices and tables

• Index

• Module Index

• Search Page

If the library is useful for your work please consider citing PyRelationAL.

@misc{pyrelational,
      title={PyRelationAL},
      author={Relation Therapeutics},
      year={2021},
      publisher = {GitHub}
      journal = {GitHub repository}
      howpublished = {\url{https://github.com/RelationRx/pyrelational}}
}